Dini, F. L., Bajraktari, G., Zara, C., Mumoli, N. & Rosa, G. M. Optimizing management of heart failure by using echo and natriuretic peptides in the outpatient unit. Adv. Exp. Med. Biol.1067, 145–159 (2018).
PubMed
Google Scholar
De Albuquerque, F. et al. Early warning systems for the management of chronic heart failure: a systematic literature review of cost-effectiveness models. Expert Rev. Pharmacoecon Outcomes Res.18, 161–175 (2018).
Google Scholar
Lippi, G. & Sanchis-Gomar, F. Global epidemiology and future trends of heart failure. AME Med. J.5 (2020).
Michas, G. et al. Heart failure in greece: the Hellenic National nutrition and health survey (HNNHS). Hellenic J. Cardiol.62, 315–317 (2021).
PubMed
Google Scholar
Buckingham, S. A. et al. Home-based versus centre-based cardiac rehabilitation: abridged Cochrane systematic review and meta-analysis. Open. Heart. 3, e000463 (2016).
PubMed
PubMed Central
Google Scholar
Grace, S. L., Kotseva, K. & Whooley, M. A. Cardiac rehabilitation: Under-Utilized globally. Curr. Cardiol. Rep.23, 118 (2021).
PubMed
Google Scholar
Stefanakis, M., Batalik, L., Antoniou, V. & Pepera, G. Safety of home-based cardiac rehabilitation: a systematic review. Heart Lung. 55, 117–126 (2022).
PubMed
Google Scholar
Piepoli, P. MF et al. 2016 European guidelines on cardiovascular disease prevention in clinical practice: the sixth joint task force of the European society of cardiology and other societies on cardiovascular disease prevention in clinical practice (constituted by representatives of 10 societies and by invited experts)Developed with the special contribution of the European Association for cardiovascular prevention & Rehabilitation (EACPR). Eur. Heart J.37, 2315–2381 (2016).
PubMed
PubMed Central
Google Scholar
Tang, L. H. et al. Patients’ preference for exercise setting and its influence on the health benefits gained from exercise-based cardiac rehabilitation. Int. J. Cardiol.232, 33–39 (2017).
PubMed
Google Scholar
Gao, L. et al. Medical and non-medical students’ knowledge, attitude and willingness towards the COVID-19 vaccine in china: a cross-sectional online survey. Hum. Vaccin. Immunother.18, 2073757 (2022).
PubMed
PubMed Central
Google Scholar
Twinamasiko, N. et al. Assessing knowledge, attitudes and practices towards COVID-19 public health preventive measures among patients at Mulago National referral hospital. Risk Manag. Healthc. Policy14, 221–230 (2021).
PubMed
PubMed Central
Google Scholar
Wang, J., Chen, L., Yu, M. & He, J. Impact of knowledge, attitude, and practice (KAP)-based rehabilitation education on the KAP of patients with intervertebral disc herniation. Ann. Palliat. Med.9, 388–393 (2020).
PubMed
Google Scholar
Yang, Z., Jia, H. & Wang, A. Predictors of home-based cardiac rehabilitation exercise adherence among patients with chronic heart failure: a theory-driven cross-sectional study. BMC Nurs.22, 415 (2023).
PubMed
PubMed Central
Google Scholar
Son, Y-J., Choi, J. & Lee, H-J. Effectiveness of nurse-led heart failure self-care education on health outcomes of heart failure patients: a systematic review and meta-analysis. Int. J. Environ. Res. Public Health. 17, 6559 (2020).
PubMed
PubMed Central
Google Scholar
Gondekar, A. et al. Knowledge, attitude, and practice of physiotherapists about cardiac rehabilitation program adherence among patients discharged from the hospital after cardiac surgery in India. Scientific World J.8825476, (2024).
Jolly, K. et al. The Birmingham rehabilitation uptake maximisation study (BRUM). Home-based compared with hospital-based cardiac rehabilitation in a multi-ethnic population: cost-effectiveness and patient adherence. Health Technol. Assess.11, 1–118 (2007).
PubMed
Google Scholar
Dankner, R. et al. A controlled intervention to increase participation in cardiac rehabilitation. Eur. J. Prev. Cardiol.22, 1121–1128 (2015).
PubMed
Google Scholar
Zhang, X., Yu, Y., Jin, O. & Zhang, L. Efficacy of novel phased health education in the management of anorectal care. Am. J. Translational Res.15, 4255 (2023).
Google Scholar
Thomas, R. J. et al. Home-Based cardiac rehabilitation: A scientific statement from the American association of cardiovascular and pulmonary rehabilitation, the American heart association, and the American college of cardiology. J. Am. Coll. Cardiol.74, 133–153 (2019).
PubMed
PubMed Central
Google Scholar
Bloom, B. S. (ed) Learning for Mastery. Instruction and Curriculum. Regional Education Laboratory for the Carolinas and Virginia, Topical Papers and Reprints, Number 11968.
Gyeltshen, K., Phuntsho, S. & Wangdi, K. Knowledge, attitude, and practice towards COVID-19 among patients attending Phuentsholing hospital, Bhutan: A Cross-Sectional study. Int. J. Environ. Res. Public. Health20 (2023).
Yang, J., Liao, Y., Hua, Q., Sun, C. & Lv, H. Knowledge, attitudes, and practices toward COVID-19: A cross-sectional study during normal management of the epidemic in China. Front. Public. Health. 10, 913478 (2022).
PubMed
PubMed Central
Google Scholar
Wu, J., Xie, Z., Xiao, Y., Wang, B. & Zhang, P. Prognostic nomogram for female patients suffering from non-metastatic Her2 positive breast cancer: A SEER-based study. Med. (Baltim).101, e30922 (2022).
Google Scholar
Yang, J. et al. Path analysis of influencing factors of depression in Middle-Aged and elderly patients with diabetes. Patient Prefer Adherence. 17, 273–280 (2023).
PubMed
PubMed Central
Google Scholar
Lublóy, Á. Medical crowdfunding in a healthcare system with universal coverage: an exploratory study. BMC Public. Health. 20, 1672 (2020).
PubMed
PubMed Central
Google Scholar
Liu, W., Wang, Y. & Wang, Z. An empirical study of continuous use behavior in virtual learning community. PLoS One. 15, e0235814 (2020).
PubMed
PubMed Central
Google Scholar
Qian, J., Zhang, W., Qu, Y., Wang, B. & Chen, M. The enactment of knowledge sharing: the roles of psychological availability and team psychological safety climate. Front. Psychol.11, 551366 (2020).
PubMed
PubMed Central
Google Scholar
Laksono, A. D., Wulandari, R. D., Rohmah, N., Rukmini, R. & Tumaji, T. Regional disparities in hospital utilisation in indonesia: a cross-sectional analysis data from the 2018 Indonesian basic health survey. BMJ Open.13, e064532 (2023).
PubMed
PubMed Central
Google Scholar
Hao, H. et al. Productivity losses due to diabetes in urban rural China. Int J. Environ. Res. Public. Health19 (2022).
Alvarez, P., Sianis, A., Brown, J., Ali, A. & Briasoulis, A. Chronic disease management in heart failure: focus on telemedicine and remote monitoring. Rev. Cardiovasc. Med.22, 403–413 (2021).
PubMed
Google Scholar
Son, Y. J., Choi, J. & Lee, H. J. Effectiveness of Nurse-Led heart failure Self-Care education on health outcomes of heart failure patients: A systematic review and Meta-Analysis. Int J. Environ. Res. Public. Health17 (2020).
Zhang, X., Yu, Y., Jin, O. & Zhang, L. Efficacy of novel phased health education in the management of anorectal care. Am. J. Transl Res.15, 4255–4261 (2023).
PubMed
PubMed Central
Google Scholar
Heidenreich, P. A. et al. 2022 AHA/ACC/HFSA guideline for the management of heart failure: A report of the American college of cardiology/american heart association joint committee on clinical practice guidelines. Circulation145, e895–e1032 (2022).
PubMed
Google Scholar
Nichols, S., McGregor, G., Breckon, J. & Ingle, L. Current insights into Exercise-based cardiac rehabilitation in patients with coronary heart disease and chronic heart failure. Int. J. Sports Med.42, 19–26 (2021).
PubMed
Google Scholar
Arfaras-Melainis, A. et al. Heart failure and sepsis: practical recommendations for the optimal management. Heart Fail. Rev.25, 183–194 (2020).
PubMed
Google Scholar
Girerd, N. et al. Practical outpatient management of worsening chronic heart failure. Eur. J. Heart Fail.24, 750–761 (2022).
PubMed
Google Scholar
Carrillo, M. A., Kroeger, A., Cardenas Sanchez, R., Diaz Monsalve, S. & Runge-Ranzinger, S. The use of mobile phones for the prevention and control of arboviral diseases: a scoping review. BMC Public. Health. 21, 110 (2021).
PubMed
PubMed Central
Google Scholar
Yu, K. et al. Modified ultrasound scalpel haemorrhoidectomy versus conventional haemorrhoidectomy for mixed haemorrhoids: a study protocol for a single-blind randomised controlled trial. Trials24, 140 (2023).
We conducted a prospective cohort at The King Chulalongkorn Memorial Hospital in Bangkok, Thailand. The study included participants aged ≥ 18 years diagnosed with EGFR-mutated recurrence or advanced-stage NSCLC. EGFR mutation testing was conducted using single gene testing Cobas® mutation test v2. or diver alteration gene panel. All participants received EGFR TKIs (1st -3rd generation) as first-line treatment, according to the provided physician. The pretreatment assessment and response evaluation were conducted as a standard practice of the institute. Demographic characteristics were obtained from the hospital’s electronic medical records. I confirm that all experiments were performed in accordance with the Declaration of Helsinki. The Institutional Review Board of the Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand, approved the study. (IRB No. 894/63 and 580/66). Written informed consent was obtained from all participants. The Bureau of Registration Administration, Ministry of Interior, Bangkok, Thailand, validated the participant’s death date.
Tumor immune microenvironment assessment
Tissue samples were collected upon the diagnosis of advanced-stage non-small cell lung cancer. CD8 Tumor-infiltrating lymphocytes (TILs) and PD-L1 were evaluated by immunohistochemistry. The interpretation was made by one pathologist (S.S.) who was blinded to clinical outcomes. TILs were assessed using immunohistochemistry staining to evaluate the expression of CD8 + T-cells according to the guidelines established by the International TILs Working Group in 2014 17. The evaluation was based on the spatial location of CD8 + TILs, intra-tumoral or stromal. Intra-tumoral CD8 + TILs were defined as CD8 + TILs with direct cell-cell contact with carcinoma cells. The results will be presented as percentages based on tumor cells17. High intra-tumoral CD8 + TILs were defined as intra-tumoral CD8 + TILs ≥ 10%18. PD-L1 was assessed using the Dako FLEX 22C3 and presented by the tumor proportion score19,20. High PD-L1 was defined as PD-L1 TPS ≥ 15% as previously demonstrated the correlation with prognosis21. The definition of inflammatory tumor microenvironment was defined as either PD-L1 TPS ≥ 15% or intra-tumoral CD8 + TILs ≥ 10% as previously reported4,18,21,22.
Blood specimen correction and HLA typing evaluation
Blood samples were collected before the participants started EGFR TKIs treatment in an EDTA tube, centrifuged, and kept at -80 °C until further process. The evaluation of HLA typing was performed by buffy coat using whole-exome sequencing technology. Briefly, DNA from the buffy coat was extracted using a Qiagen blood mini kit following manufacturer protocol. Library preparation was proceeded using SureSelectXT V6 + UTR library prep kit (Illumina, San Diego, CA, USA). The sequencing was conducted using NovoSeq to generate 150 bp paired-end reads at Macrogen Inc. (Seoul, Korea). We analyzed data through bcbio-nextgen version v1.2.923 with target sequences of approximately 90 Mb. Unmapped BAM was generated from Fastq raw data, aligned with hg38 reference using BWA version 0.7.1724, and processed by using the GATK best practice pipeline through Genome Analysis Toolkit recommendation (GATK version 4.1.0.0 including MarkDuplicates, base quality score recalibration, indel realignment, duplicated removal. We identify high polymorphic HLAs using OptiType algorithm25 which shown high accuracy26. The HLA-A was classified into interested and non-interested subtypes based on the binding affinity of the EGFR mutation subtype, which had been previously reported15. The higher binding affinity of the HLA-A subtype represented potential neoantigen. The interested HLA-A was also correlated with favorable prognostic outcomes in resectable NSCLC15. For EGFR L858R alteration, interested HLA-A subtypes were HLA-A*30:01, HLA-A*31:01, HLA-A*33:01, HLA-A*33:03, HLA-A*34:01, HLA-A*66:02, HLA-A*68:01, HLA-A*68:03, HLA-A*68:04, and HLA-A*68:05. While EGFR exon 19 deletion, interested HLA-A subtypes were HLA-A*03:01, HLA-A*03:02, HLA-A*11:01, HLA-A*30:01, HLA-A*34:02, HLA-A*68:01. The presence of one allele of interested HLA-A was considered positive for interested HLA-A.
Sample size calculation
The author was calculated based on Dimou A., et al., who reported that the interesting HLA-A alleles, i.e., HLA-A*11:01, HLA-A*24:02, HLA-A*02:03, HLA-A*33:03, and HLA-A*02:07, exhibited greater binding efficacy to either EGFR L858R or exon 19 deletion peptides. The prevalence of HLA-A*11:01, HLA-A*24:02, HLA-A*02:03, HLA-A*33:03, and HLA-A*02:07 of the Thai population was 26%, 11%, 11%, 11%, and 8%, respectively as previously reported by Satapornpong et al.16. The prevalence of the inflammatory TIME reported by Matsumoto et al. was 13.5%4. We proposed a hypothesis that an interested HLA-A results in a four-fold higher frequency of inflamed TIME compared to uninterested HLA-A subtypes. Using a proportion sample size calculation27to achieve a Type 1 error rate of 5% and a power of 80%, the sample size was calculated to be 74, without continuity correction.
Statistical analysis
Categorical data was analyzed using the Chi-square or Fisher exact test. Continuous data was analyzed using the Mann-Whitney test. The correlation between the HLA-A and either inflammatory TIME or intra-tumoral CD8 TILs was calculated using the Chi-square test. Progression-free survival (PFS) was defined by the time of initiation EGFR-TKIs treatment to the date of objective disease progression or death from any cause. Overall survival (OS) was determined by the time of initiation of EGFR-TKIs treatment to the date of death from any cause. The data was censored on December 31, 2023, for alive or non-progressive disease participants. Multivariate analyses of clinical factors, HLA-A subtype, and tumor immune microenvironment expression level were performed using a Cox proportional hazards model. The Kaplan-Meier method was used to evaluate survival, and the log-rank test was used to evaluate the significance of the difference between groups. The significance level was defined as p-value < 0.05. Statistical analysis was performed using SPSS version 29.0.
This section outlines the deception detection methodology proposed in this study. It discusses the traditional SQM metrics, as well as the MuSD and MuSDA metrics and their threshold calculation methods. Additionally, the computation process of the WMA-BC algorithm is also examined.
Spoofing detection process
The spoofing detection process (Fig. 2) follows a sequential workflow: First, the GNSS receiver demodulates the intermediate frequency (IF) signal. Then, multiple correlators process this signal to generate correlator outputs, which are used to calculate two key metrics—MuSDA and MuSD. These metrics undergo further processing through the WMA-BC algorithm for enhanced detection capability. Finally, the processed metric values (M_{x}) are compared against thresholds (theta_{x}) for spoofing signals to determine whether the received signal is authentic or spoofed.
Fig. 2
Flowchart of WMA-BC for GNSS spoofing detection.
Expressions of the monitoring metrics
Monitoring metrics utilize the correlator output parameter with different composition methods to detect spoofing signals. To understand the metrics, the statistical characteristics must be known. The mean value and noise variance of the metrics can be obtained through calculations, and differences in the signal-to-noise ratio, correlation integral time, and correlator positions can change the noise variance of the metrics. We considered six SQM metrics for comparison: the ELP, ratio, delta, double delta, slope, and double slope metrics. The detailed derivation of the noise variance of each metric is presented in the Appendix.
Table 1 summarizes the definitions and statistical characteristics of the SQM metrics. ({{I_{E} } mathord{left/ {vphantom {{I_{E} } {Q_{E} }}} right. kern-0pt} {Q_{E} }}) and ({{I_{L} } mathord{left/ {vphantom {{I_{L} } {Q_{L} }}} right. kern-0pt} {Q_{L} }}) are the values of the early/late correlator in the in-phase/quadrature correlators, where (E) and (L) are 0.5 and -0.5, respectively. (I_{0}) denotes the output value of the maximum correlator; (I_{{d_{1} }}), (I_{{d_{2} }}), (I_{{ – d_{1} }}) and (I_{{ – d_{2} }}) are the output values of the additional correlators, where the negative sign represents early and no negative sign represents late, (d) denotes the spacing between the correlators and the maximum correlator, and the numbers are used as identifiers. The unit of (d) is chips.
Table 1 Definitions and theoretical statistics of the SQM monitoring metrics.
MuSD and MuSDA metrics
This section describes the NeSD, MiSD, FaSD, MuSD, and MuSDA metrics. Correlators at different locations have distinct offset detection advantages. NeSD, MiSD, and FaSD have complementary properties because the correlators used are at specific locations. However, relying solely on the correlators used by NeSD, MiSD, or FaSD to obtain spoofing detection results is unreliable. MuSD and MuSDA effectively utilize all the different correlators used by NeSD, MiSD, and FaSD by aggregating slope information from correlators at different offsets. This method leverages this diversity to build a more comprehensive signal integrity profile, efficiently increasing the spoofing detection range and preventing, slope, ratio, and ELP from effectively detecting only significant changes at the top or both sides of the correlation peak.
The correlators needed to construct the metrics are shown in Fig. 3. The MiSD correlator (d_{2} = 0.5) chips, which is the (E – L) correlator of the receiver, and (d_{0} = 0) chips, which is the prompt correlator of the receiver. The NeSD and MiSD correlators are located on either side of the (E – L) correlator. The correlator spacing is (d_{1} > d_{2} > d_{3} > d_{0}) , which is determined based on the offset detection advantages of using correlators at different locations.
Fig. 3
Correlator locations for NeSD, MiSD, FaSD, MuSDA and MuSD. The blue dots indicate added correlators, and the red dots indicate the original E, L, P correlators of the receiver.
The (d_{3}) correlator of NeSD is effective for monitoring the ACF near-point distortion by very small code phase difference spoofing and low power spoofing due to its proximity to the prompt correlator. The NeSD can be expressed as
Intermediate code phase difference spoofing causes ACF distortion near the correlator, reducing the spoofing detection capability of NeSD for middle and far distortion points. MiSD utilizes the receiver’s original (E – L) correlator to supplement the NeSD performance. The metric is defined as
Due to correlation limitations, NeSD and MiSD have difficulty detecting small ACF distortions at far points. Thus, to prevent spoofing leakage detection, the far point correlator (d_{1}) is deployed beyond the (E – L) correlator, and the FaSD metric is defined as follows:
MuSD is a joint decision metric that is not directly obtained by the correlators but is jointly determined by the NeSD, MiSD, and FaSD results through logical association operations. The MuSD metric is expressed as follows:
where (H_{SDN}), (H_{SDI}), and (H_{SDF}) are the NeSD, MiSD, and FaSD decision results, respectively.
The MuSDA metric is derived from the receiver’s E, L, and P correlators, as well as additional correlators, using the mean-value difference method, and the MuSDA metric is defined as
MuSD and MuSDA use the same correlator and can therefore be used simultaneously to detect spoofing.
Theoretical thresholds and decision rules for metrics
The threshold for the metrics can be adaptively calculated based on the desired false alarm rate and the statistical characteristics. For satellite navigation signals, the hypothesis testing theory of signal processing is used to identify spoofing signals, with a null hypothesis (H_{0}) indicating that no spoofing signal exists and an alternative hypothesis (H_{1}) indicating that a spoofing signal exists. Assuming that the probability density function of the noise in the case of (H_{0}) follows a normal distribution with mean (mu_{x}) and standard deviation (delta_{x}), the false alarm rate (P_{fa}) is expressed as
where (erfc^{ – 1} (x)) is the inverse function of (erfc).
The false alarm rate can be flexibly adjusted according to the specific requirements of different application scenarios. For instance, high-risk scenarios may tolerate a slightly higher false alarm rate to ensure critical events are not missed, whereas low-risk applications require stricter control over the false alarm rate to avoid unnecessary disruptions.
In the spoofing monitoring process, the decision is divided based on the results of the comparison between the metric measurement and its threshold. The discriminant is as follows:
If the metric measurements exceed the thresholds, there is a spoofing signal; otherwise, there is no spoofing signal.
Weighted moving average bias correction
Noise, spoofing signals and other interference sources all cause transient or short-term fluctuations in metric data. When the spoofed signal operates in the frequency unlocking mode, the relative carrier phases of the real and spoofed signals change over time, leading to significant oscillations in the monitoring metrics and causing unnecessary false alarms33. To reduce the influence of noise interference, we propose the weighted moving average bias correction algorithm, which can be applied to metric data. This approach considers recent data obtained over time, smooths the curve of the monitoring data, reduces the influence of random interference, and improves the robustness and detection performance of the metrics. This subsection describes the computational process and analyzes the simulation results obtained with this method. The traditional weighted moving average algorithm (WMA) expression is
$$P_{t} = beta P_{t – 1} + (1 – beta )M_{t}$$
(20)
where (P_{t}) and (P_{t – 1}) are the predicted values of the monitoring data at moments (t) and (t – 1) , respectively.(M_{t}) is the measured value at moment (t), where (beta) represents the rate of the decay weights, and its expression is
$$beta = 1 – frac{1}{{T_{c} }}$$
(21)
The moving window size in the WMA-BC algorithm is directly related to the receiver’s PIT, as (T_{c}) represents the minimum time interval over which coherent signal accumulation occurs. Therefore, we set the window size equal to the PIT.
During the receiver’s operation, we adopt an adaptive PIT adjustment strategy based on the scene type to optimize system performance. This mechanism determines the current scene type (low-speed/static or dynamic) by analyzing the Doppler rate (Delta {text{f}}) of change and dynamically adjusts the PIT. When the Doppler rate of change is less than or equal to 2 Hz/s, the system identifies the scene as low-speed/static, and the PIT is set to 10 ms to save computational resources and improve response speed. When the Doppler rate of change exceeds 2 Hz/s, the system identifies the scene as dynamic, and the PIT is adjusted to 100 ms to accommodate environments with large frequency fluctuations. The PIT can be expressed as:
The WMA algorithm smooths ACF data by averaging past observations, giving more weight to recent data. While this reduces jitter, it can introduce bias due to small initial values. To address this, we propose the WMA-BC algorithm, which adds a bias correction step to reduce the discrepancy between smoothed and actual values, improving prediction accuracy. The WMA-BC algorithm is as follows:
where (P_{{biased_{t} }}) is the weighted moving average bias correction, (P_{t}) is the predicted weighted moving average.
Experimental results and discussion
Performance of the WMA-BC algorithm
To evaluate the performance of the WMA-BC algorithm, we conducted experiments on the TEXBAT dataset. In the TEXBAT dataset, Cases 2–8 are examples of synchronized spoofing intrusions, while Case 1 is an example of spoofing switching34.
In Case 2, the spoofed signal has a higher power (+ 10 dB) than does the authentic signal, and the spoofers operate in frequency-unlocked mode (the carrier phase difference between the spoofed and authentic signals is not fixed). Case 3 differs from Case 2 in that the spoofed signal operates in frequency-locked mode (the carrier phase difference between the spoofed and authentic signals is fixed), and the power is reduced from 10 dB to 1.3 dB.
We compared the spoofing detection rates of SQM metrics using WMA-BC, WMA, MA-based, and MV-based algorithms, as well as MuSDA and MuSD metrics in Case 2 (Fig. 4(a)) and Case 3 (Fig. 4(b)). The spoofing detection times ranging from 60 to 300 s and the predetection integration time (T_{c} = 100ms), and (P_{fa} le 10^{ – 7}). The detection rate is defined as
$${text{detection rate}} = frac{{text{Samples that exceed the detction threshold}}}{{text{Samples in which deception exists}}}$$
(25)
Fig. 4
(a) Detection rates of the SQM MuSDA and MuSD metrics with the WMA-BC algorithm. (Case 2,(T_{c} = 100ms), and (P_{fa} le 10^{ – 7}) ). (b) Detection rates of the SQM MuSDA and MuSD metrics with the WMA-BC algorithm. (Case 3,(T_{c} = 100ms), and (P_{fa} le 10^{ – 7}) ).
In Case 2, the detection rates of the metrics obtained based on the WMA-BC algorithm are all improved, but the effect differs for different metrics, with the detection rates of the slope, ratio, MuSDA and MuSD metrics significantly improved by more than 70%. As the detection rates of the double slope, delta, double delta, and ELP metrics were originally close to 0, the improvement in the detection rate was limited. Case 3 shows results similar to those of Case 2; the metrics obtained based on the WMA-BC algorithm have higher detection rates, and the detection rates of the metrics improve by approximately 22% to 53%.
Experimental data analysis shows significant performance differences among the four methods in the spoofing detection task. In Case 2 testing, the WMA-BC method performed the best, achieving detection rates of 100% and 81.4% for the MuSDA and slope metrics, respectively, which represents an improvement of 4.8% and 38.3% compared to the WMA algorithm. In comparison, the MA method only reached 66.3% and 26.2%, while the MV method achieved 40.4% and 32.5%. Notably, the MV method completely failed on the ratio metric (0% detection rate), whereas WMA-BC maintained an effective detection rate of 87.1%, which is an improvement of 15.1% compared to WMA.
Further analysis of the Case 3 data reveals that the advantage of WMA-BC is even more pronounced on the slope metric, where its detection rate was more than 30% higher than both the MA and MV methods, with an improvement of about 5% to 20% compared to the WMA method. This method also maintained stable performance on the double slope, delta, MuSDA, and MuSD metrics, demonstrating strong overall performance. In contrast, while the MA method performed reasonably well on the ratio and MuSD metrics, it achieved only a 14.1% detection rate on the delta metric, showing clear performance limitations. The MV method exhibited unbalanced characteristics: it performed excellently on the double slope (63.7%), double delta (63.7%), and ELP (52.6%) metrics but underperformed on key metrics such as ratio (26.2%), MuSDA (5.9%), and MuSD (23.4%).
Overall, WMA-BC demonstrated clear advantages over WMA, MV and MA in terms of performance. The superior performance of WMA-BC is primarily attributed to its weighted computation mechanism, which dynamically adjusts weights over time, effectively enhancing feature differentiation. In contrast, the MA method, due to its simple mean calculation, is prone to losing crucial temporal information. The MV method, which treats all data points within the window equally, cannot implement variance calculations that incorporate features such as exponential decay weights, which are more suited to time-series characteristics, resulting in insufficient performance.
Quantitative analysis demonstrates that WMA-BC significantly enhances the discriminative capability of embedded GNSS systems, achieving more than a 70% improvement in detection rate for key signal metrics. For instance, in Case 2, the Slope metric shows a remarkable increase from 3.6% (raw metric) to 81.4% with WMA-BC, outperforming the MA approach at 26.2% and the MV method at 32.5%.
Despite the computational burden associated with multi-correlator architectures, such as MuSDA/MuSD, which require around 750 M floating-point operations per second (FLOPs), including an overhead of 200 M FLOPs due to additional correlators—WMA-BC introduces only a minimal overhead of 0.50 M FLOPs (just 0.07% of the baseline), along with an additional 48 kB of RAM usage (Table 2).
Table 2 FLOPs and RAM overheads of three algorithms under multi-correlator metrics.
This efficiency makes WMA-BC, along with MuSD and MuSDA metrics, highly suitable for deployment on resource-constrained embedded platforms. For example, when implemented on the TMS320C6748 processor (clocked at 456 MHz with a peak performance of 3648 MFLOPs and a typical power consumption of 1.1W), the added computational load is relatively small, representing less than 5.5% of the total processing capacity. Moreover, WMA-BC maintains linear time complexity O(n), ensuring robust scalability for real-time applications—unlike MV, which has O(n2) complexity and incurs a 13 M FLOP overhead, approximately 72 times greater than that of WMA-BC.
These findings underscore WMA-BC’s unique ability to balance robust spoofing detection with stringent resource efficiency, fulfilling the demanding requirements of real-time GNSS spoofing detection systems in resource-constrained environments.
We calculated and plotted the receiver operating characteristic curves for Case 2 (Fig. 5(a)) and Case 3 (Fig. 5(b)), and performed statistical analysis (Table 3). In the experiments, (P_{d}) and (P_{fa}) were measured by continuously decreasing the metric thresholds. We evaluated the AUC, which is an important parameter for detection performance.
Fig. 5
(a) Comparison of ROC curves for different metrics (Case 2). The solid lines with circles represent curves obtained without the WMA-BC algorithm, the dashed lines with asterisks represent curves obtained based on the WMA-BC algorithm, and the dot with a triangle indicates the curve obtained based on the WMA algorithm. (b) Comparison of ROC curves for different metrics (Case 3). The solid lines with circles represent curves obtained without the WMA-BC algorithm, the dashed lines with asterisks represent curves obtained based on the WMA-BC algorithm, and the dot with a triangle indicates the curve obtained based on the WMA algorithm.
Table 3 The summary of AUC of ROC curves for different metrics.
In Case 2, the AUC values of the metrics obtained based on the WMA-BC algorithm are markedly larger than those of the metrics obtained without the WMA-BC algorithm and those using the WMA algorithm. The AUC area with the WMA-BC algorithm increased by 0.01 to 0.197 compared to the WMA algorithm, and increased by 0.043 to 0.376 compared to the original metrics. Additionally, the MuSDA metric obtained based on the WMA-BC algorithm has an AUC equal to 1, demonstrating a 100% detection rate with no false alarms. In Case 3, the AUC area of the metrics combined with the WMA-BC algorithm increased by 0.01 to 0.072 compared to the WMA algorithm, and the AUC increased by 0.038 to 0.207 compared to the original metrics, which also indicates that these metrics achieved stronger detection capabilities.
In summary, the WMA-BC algorithm-based metrics achieve enhanced spoofing detection rates and superior performance with minimal computational overhead. The GNSS receiver thereby attains improved ROC performance—regardless of whether spoofing signals operate in frequency-locked or frequency-unlock modes, and irrespective of spoofing signal power being higher than or approximately equal to authentic signals.
Spoofing detection experiments with different code phase offsets and carrier phase offsets
To examine the detection performance of various metrics in synchronized spoofing against different code phase offsets and carrier phase offsets (based on the WMA-BC algorithm), we perform experiments by simulating a GPS satellite with the following signal simulation parameters: ({C mathord{left/ {vphantom {C {N_{0} }}} right. kern-0pt} {N_{0} }}) is 45 dB, the C/A code phase difference (mathop tau limits^{ sim } = left| {tau_{s} – tau_{a} } right|) between the authentic signal and the spoofing signal ranges from 0 to 1 chip, with a step size of 0.005 chips, and carrier phase difference (mathop {theta_{s} }limits^{ sim } = left| {theta_{s} – theta_{a} } right|) ranges from 0 to 2π, with a step size of 0.1π, for a total of 4221 grid experiments. Due to the high similarity between spoofing signals and authentic signals in the experiment, and considering the tracking stability of the receiver and the timeliness of spoofing detection, (T_{c}) is 10 ms. The correlators used to obtain the MuSD and MuSDA metrics in the experiments are (d_{1} = 0.9) chips, (d_{2} = 0.5) chips, and (d_{3} = 0.1) chips.
The spoofing detection rates at the experimental grid points are demonstrated in Fig. 6, where the grid color indicates the spoofing detection probability. Each grid represents the detection rate in one experiment, and (P_{fa} le 10^{ – 7}), which can effectively reflect the detection sensitivity of the metrics obtained in each experiment. The detection rates of some metrics decrease in cases with long intrusion times, such as the slope, double slope, delta, and double delta metrics. This is because in the early stage of the spoofing attack, the output values of the early and late correlators change or differ significantly, and the detection difficulty is small. However, in the spoofing attack of the middle or late stages, the change in the output values of the early and late correlators decreases, and the detection difficulty increases. However, the MuSD and MuSDA metrics are obtained using multiple correlators, which can monitor small fluctuations in bilateral slopes at multiple points simultaneously, ensuring high sensitivity and detection rates.
Fig. 6
Detection rate of each metric in different code phase shift and carrier phase shift spoofing experiments. The code phase offset (mathop tau limits^{ sim }) ranges from 0 to 1 chip, and the carrier phase shift (mathop {theta_{s} }limits^{ sim }) ranges from 0 to 2π ((P_{fa} le 10^{ – 7})).
To evaluate the detection performance of the metric more objectively, we evaluated the detection coverage of each metric. The detection coverage is the ratio of the detectable area to the total area in a certain detection region. This value is a more comprehensive reflection of the performance of the metrics. The result of each experiment is 1 unit, and the total number of units is 4221. The detection coverage is defined as
$${text{detection coverage}} = frac{{text{Detectable area }}}{{text{Total area}}}$$
(26)
The detectable area in (26) is the sum of the detectable grid points, and we set a minimum acceptable detection rate of (P_{{d_{min } }}). If (P_{d} le P_{{d_{min } }}), the grid is undetectable, and its grid value is recorded as 0; otherwise, the grid value is set as 1.
Figure 7 shows the detection coverage of each metric when the minimum acceptable detection rate is set to 0.8. The yellow region in the figure represents the detectable region ((P_{d} ge 80%)), and the blue region represents the undetectable region. For both the slope and delta metrics with two correlators and the double slope and double delta metrics with four correlators, the detectable area is smaller than that of the other metrics, and the detection coverage is less than 60%. For the ELP and ratio metrics, undetectable grids are found at their edges or at the center in more places. In contrast, the MuSD and MuSDA metrics have mainly detectable areas, except for the undetectable areas at the edges of (mathop tau limits^{ sim } le 0.055) and (left{ {begin{array}{*{20}c} { , 0 le mathop {theta_{s} }limits^{ sim } le 0.6pi } \ {1.4pi le mathop {theta_{s} }limits^{ sim } le 2pi } \ end{array} } right.). The existence of this blind spot arises from the fact that when the code/carrier phase shift of the deception signal is extremely small, its impact on the correlation peak of the real signal is negligible, typically manifesting as a very weak “boost” or “distortion.”
Fig. 7
Detectable region of each metric in different code phase shift (mathop tau limits^{ sim }) and carrier phase shift (mathop {theta_{s} }limits^{ sim }) spoofing experiments. ((P_{min } = 80%)).
Under the stable operation of the receiver, the receiver experiences fluctuations such as thermal noise and quantization noise, which are very similar to the changes caused by spoofing signals. The adaptive threshold (theta_{x}) we set has a statistical fluctuation offset range given by
When the spoofing signal is extremely similar to the true signal, the metric shift (Delta x) caused by the spoofing signal is very small, and the subtle changes induced by the spoofing signal are easily masked by the inherent noise. The expression is:
$$Delta x = left| {M_{spoofing} – mu_{x} } right|$$
(28)
where (M_{spoofing}) is the measured metric’ value when spoofing is present. Its shift mainly comes from the minor distortions of the signal generator and the additional noise introduced by the spoofing signal itself. When the metric shift (Delta x) caused by the spoofing signal is less than or equal to the statistical fluctuation range (offset) , the receiver cannot effectively detect the spoofing, leading to a detection blind spot.
The detection coverage of the eight detection metrics is summarized (Fig. 8), and the performance of the metrics can be ranked as follows: delta (26.4%) < slope (37.4%) < double delta (52.4%) < double slope (58.3%) < ELP (61.1%) < ratio (73.3%) < MuSDA (95.8%) < MuSD (96.1%). MuSD has the highest detection coverage of 96.1%, and MuSDA has a slightly lower detection coverage than the MuSD metric, with a value of 95.8%, which is approximately 22% to 69% higher than that of the other metrics. These results show that MuSDA and MuSD possess smaller blind zones and outperform the other metrics in terms of code phase offset and carrier phase offset detection.
Fig. 8
Detection coverage of different metrics ((P_{min } = 80%)).
We evaluated the performance of MuSD under different correlator spacing combinations (Table 4). The experimental results show that as (d_{1}) decreases and (d_{3}) increases (with the receiver’s inherent correlator spacing (d_{2} = 0.5)), the detection coverage of MuSD decreases significantly. In this experiment, the configuration (d_{1} = 0.9), (d_{2} = 0.5), and (d_{3} = 0.1) achieved the highest detection coverage (95.8%). This result indicates that this configuration effectively ensures high detection coverage of MuSD across different code phase offset and carrier offset experiments, demonstrating the strongest robustness.
Table 4 The detection coverage of the MuSD metric under different combinations of correlator spacings.
Test with the TEXBAT dataset
To further validate the performance of the metrics (based on the WMA-BC algorithm), we used seven spoofing intrusion cases from the TEXBAT dataset as tests. The battery can be considered the data component of an evolving standard meant to define the notion of spoof resistance for civil GPS receivers. According to this standard, successful detection of or imperviousness to all spoofing attacks in TEXBAT, or a future version thereof, could be considered sufficient to certify a civil GPS receiver as spoof resistant34. It includes dynamic, static, power matching, carrier/code phase matching, and other scenarios (Table 5), among which the challenge of spoofing detection on a dynamic platform is to distinguish spoofing effects from natural fading and multipath.
Table 5 Summary of the TEXBAT dataset.
We detected the signals from 60 to 300 s for each case (240 s in total) and selected the period from 120 to 300 s (spoofing intrusion phase) to calculate the spoofing detection rate. The PIT was set to (T_{c} = 100ms), and the false alarm rate (P_{fa} le 10^{ – 7}).
Through experiments based on the TEXBAT dataset, we visualized the detection rate for each metric (Fig. 9) and performed statistical analysis (Table 6). This result reflects the detection effectiveness in defending against deceptive intrusions. Case 2 is a time-specific attack. The detection rates of the slope, ratio, MuSDA, and MuSD metrics are relatively high, reaching greater than 80%, with MuSDA and MuSD reaching 100%. In contrast, the detection rates of the double slope, delta, double delta, and ELP metrics are not more than 25%. Case 3 is the same as Case 2 except that the power difference between the spoofed and authentic signals is reduced from 10 dB to 1.3 dB, and the spoofers operate in frequency-locked mode. The frequency-locked mode increases the change in the correlator detection value, which is favorable for spoofing detection. The detection rates of the double slope, delta, double delta, and ELP metrics are improved in this case, and the ratio, MuSDA, and MuSD metrics maintain high detection rates of 81.9%, 85.2%, and 94.4%, respectively. Case 4 is the same as Case 3 except that the power difference between the spoofed and authentic signals is reduced (from 1.3 dB to 0.4 dB), and the spoofed signals are position offset-type spoofs. Compared to the results in Case 3, the detection rates of the metrics decrease, while the MuSDA and MuSD metrics still maintain high detection rates of 92.9% and 99.1%, respectively. Case 5 is similar to Case 2, except that the receiver platform is changed from static to dynamic, and obvious changes in the power and phase values occur, making spoofing detection more difficult. The detection rates of the double slope, delta, and double delta metrics are close to 0 in this case. The detection rates of the slope and ratio metrics decrease to 10.3% and 48.7%, respectively, while the MuSDA and MuSD metrics maintain high detection rates of 97.6% and 99.9%, respectively. Case 6 is similar to Case 4, except that the receiver platform is changed from static to dynamic. In Case 6, the detection rates of the various metrics show different degrees of change, with the slope, double slope, delta, double delta, ELP, ratio, MuSDA, and MuSD metrics obtaining detection rates of 20%, 55.3%, 22.5%, 53.6%, 4.9%, 63.7%, 76.1%, and 98.8%, respectively. Case 7 is similar to Case 3, except that a carrier phase alignment strategy is implemented for the spoofed signals. In this case, the delta metric has a detection rate of 0%, the ELP metric has a detection rate of only 1.8%, the slope, double slope, and double delta metrics have detection rates between 50 and 51%, and the MuSDA and MuSD metrics have detection rates of 62.9% and 63.8%, respectively. In Case 8, zero-delay security code estimation and replay attacks are used. In this case, compared to those of Case 7, the double slope and ELP metrics still perform poorly, with detection rates of approximately 0, and the detection rates of the slope and ratio metrics decrease by 13.6% and 13.1%, respectively. The detection rates of the double slope, double delta, MuSDA, and MuSD metrics remain approximately unchanged, with MuSD showing the best detection rate of 63.3%.
Fig. 9
Detection rates of the different metrics based on the TEXBAT dataset (Cases 2 to 8). (T_{c} = 100ms),(P_{fa} le 10^{ – 7}).
Table 6 Summary of detection rates for different metrics based on the TEXBAT dataset.
Overall, among the eight metrics considered in the experiments, the MuSD and MuSDA obtained the best detection performance, showing the highest detection rates in all the experiments. The ratio and slope metrics showed the next best detection performance; although their detection capability was not as good as that of the MuSD and MuSDA metrics, they obtained good detection rates in all the cases. In contrast, the double delta, double slope, delta, and ELP metrics performed poorly, with detection rates close to 0 in some cases. The reason is that MuSD and MuSDA exploit advantages including the offset detection capability of complementary correlators, which comprehensively improves the detection capability of time-type and location-type spoofing signals, such as phase shifts, power suppression, and Earth-centered Earth-fixed coordinate deviations. It also solves the problem in that other metrics are not effective in detecting highly similar spoofing (the code phase, the carrier phase, and power are all very close), improving the spoof detection ability.
ISLAMABAD – The Securities and Exchange Commission of Pakistan (SECP) has introduced a new category of mutual funds titled “Infrastructure Schemes” under the framework of open-end collective investment schemes. This initiative represents a significant step towards strengthening the role of capital markets in channelling long-term savings into infrastructure development.
The proposal for creating a distinct category was initially presented at the Mutual Fund Focus Group Session 2025, where it was identified as a key milestone under the Fund Management Department’s Roadmap 2025–26. Extensive consultations were subsequently held with the Mutual Funds Association of Pakistan (MUFAP) and other stakeholders to refine the framework. The final structure reflects both industry feedback and SECP’s commitment to ensuring regulatory clarity, investor protection, and alignment with national development priorities.
Pakistan faces an urgent requirement to expand and modernise its infrastructure, with financing needs estimated at nearly $15 billion annually. Current infrastructure spending remains significantly below international benchmarks, amounting to just 2.1 percent of GDP compared to the global standard of 8–10 percent. By introducing a dedicated regulatory category, the commission seeks to provide stronger visibility to infrastructure-focused mutual funds, while offering investors a transparent and well-structured avenue for participation in projects of national significance.
Under the framework, Asset Management Companies (AMCs) may categorise infrastructure schemes as equity, debt, or hybrid funds depending on their investment focus. Eligible sectors include energy, transport, logistics, water, sanitation, communication, and a wide range of social and commercial infrastructure such as hospitals, educational institutions, industrial parks, affordable housing, and tourism facilities. This broadened scope is intended to mobilise both retail and institutional investors towards ventures that directly contribute to Pakistan’s development agenda.
To promote investor confidence, the framework prescribes minimum fund sizes of Rs100 million for perpetual schemes and at the close of the subscription period for closed-end schemes. AMCs will be required to invest a minimum seed capital of Rs25 million in closed-end schemes with maturity exceeding three years, ensuring alignment of interest between managers and investors. Closed-end schemes may also offer periodic subscription and redemption windows after one year, subject to conditions clearly set out in the offering documents. The framework provides flexibility in relation to Net Asset Value (NAV) disclosure for closed-end infrastructure schemes, requiring disclosure at intervals not exceeding one month as specified in the constitutive documents. In addition, schemes must maintain at least 70 percent of net assets invested in infrastructure securities on a quarterly basis, with any shortfall to be regularised within three months.
A transparent fee structure has also been introduced. Management fees are capped at three percent per annum for equity schemes and 1.5 percent for debt schemes, while hybrid schemes will follow a weighted average based on asset allocation. No sales load will be permitted, though contingent load may apply in the case of early redemption under closed-end schemes. By establishing this dedicated category, SECP seeks to bridge Pakistan’s infrastructure financing gap through long-term domestic savings, while ensuring strong investor safeguards. The initiative reaffirms SECP’s commitment to fostering sustainable growth and deepening capital markets as a vehicle for economic development. The circular is available on SECP’s website.
FROM classrooms in Lahore to farms in Sindh, artificial intelligence (AI) could soon be as common as smartphones if Pakistan uses it wisely. Large Language Models (LLMs) represent one of the most powerful applications of AI.
Many still think of them only as chat-bots, but their potential extends far beyond that. They can reshape Pakistan’s most important sectors by improving efficiency, expanding access to services, and driving innovation. If deployed thoughtfully with due planning, they could bring meaningful change to education, healthcare, agri- culture, finance and public services.
A study in the Pakistan Journal of Life and Social Sciences found that teachers already see the potential for using LLMs to support the learning of English as a Foreign Language (EFL). In classrooms, LLMs can help teachers prepare lesson materials, create practice questions, and give students personalised feedback. English language learners in rural areas, for example, could get additional practice through AI tools after their school hours.
The study noted that simple chat and questioning tools were the most widely used by teachers, while more complex features, like advanced content creation, were less common.
This suggests that straightforward, targeted AI tools could have the fastest impact. Still, teachers must be trained to review and refine AI-generated material to ensure accuracy and relevance. Without this oversight, there is a risk of misinformation or over-reliance on AI.
Similar opportunities exist in health-care, where AI can bridge critical gaps in access to services and improve patient outcomes.
Pakistan has an acute shortage of doctors, particularly in rural and under-served areas. LLM-powered tools could support nurses and other healthcare workers by offering quick symptom checks, translating medical information into local languages, and guiding patients through basic health steps. However, these tools must be carefully tested to protect patient privacy, ensure medical accuracy, and maintain public trust.
Further, AI can transform the vital sector of agriculture by equipping farmers with timely, localised information, including weather forecasts, crop-care advice, and early warnings about plant diseases, all in their own language. This could improve decision-making and boost yields.
Beyond the fields, AI has the potential to modernise Pakistan’s financial system, making banking more secure, accessible and user-friendly. In banking and financial services, LLMs can help detect fraud, improve customer services and analyse risks. Besides, they can simplify complex financial terms for the public, helping more people make informed decisions.
However, realising these benefits across sectors will require overcoming significant challenges, ranging from language biases to limited digital infrastructure.
Research has found that LLMs often behave differently in local languages compared to English. This can lead to unintended biases in how information is presented.
Other such challenges include limited internet access in rural areas, the high cost of AI tools, and the danger of over-
reliance on AI without human oversight. These risks must be addressed before rolling out large-scale AI initiatives.
If Pakistan invests in the right tools, trains its workforce, and ensures fair access, AI could help address some of the most pressing challenges that the country is facing. The technology is ready. It is time for Pakistan to get ready as well.
ISLAMABAD: Pakistan’s economic recovery has been stronger than expected, yet the country has failed to seize an opportunity to put its recovery on a sustainable path due to the absence of bold and long-lasting reforms, according to the Washington-based Institute of International Finance (IIF).
In a special report, the IIF noted that while Pakistan has successfully rebuilt its economic buffers and secured financing, the gains will likely prove short-lived without comprehensive structural reforms, particularly in tax broadening, privatisation, and the resolution of circular debt.
The report emphasised that Pakistan has made little headway in these critical areas, particularly privatisation and energy-sector restructuring, with circular debt still unresolved. The IIF warned that these unresolved issues pose a significant risk to Pakistan’s economic outlook for FY26.
Notably, inflation has decreased significantly, allowing the State Bank of Pakistan (SBP) to cut its policy rate to 11pc since the easing cycle began in June 2024.
In addition, Pakistan posted its first current account surplus (0.5pc of GDP) since FY11, along with the highest primary balance surplus (2.4pc of GDP) in over two decades in FY25. These developments have resulted in sustained multilateral and bilateral support and improved credit ratings.
However, the IIF highlighted that despite these positive headlines, the economic situation is not as promising as it may seem. Geopolitical tensions, both regional and global, pose significant challenges for FY26, while domestic political instability, though subsiding, remains fragile. The relationship between the military establishment and the opposition PTI party remains tenuous, adding to the uncertainty.
Highlights structural weaknesses in tax broadening, privatisation and energy sector, undermining long-term stability
While fiscal and external buffers accumulated in FY24/25 have provided some relief, they remain limited. The $5bn increase in reserve assets has boosted the country’s import coverage to just 2.4 months, while the primary balance surplus has led to a slight reduction in total public sector debt, which remains high at around 67pc of GDP. These figures suggest that while short-term stability has been achieved, long-term sustainability remains uncertain.
The IIF also pointed out that the recent trade agreement with the United States, Pakistan’s largest export partner, could provide some support to the textile industry, though the benefits are expected to be modest. Agriculture, which accounts for nearly a quarter of GDP and employs 40pc of the workforce, will likely remain sluggish. The kharif season, covering key crops such as rice, sugarcane, cotton, and maise, has faced early water shortages followed by heavy monsoon rains, which could weigh heavily on harvests in the first half of FY26.
Furthermore, deadly flash floods have exacerbated the challenges, plunging Pakistan into its second major flooding crisis in three years. This could have serious implications for growth, as well as for the country’s external and fiscal balances.
Inflation, while improving, remains a concern. A sharp rise in food prices, caused by the floods, led to a 2.9pc month-on-month increase in headline inflation in July, the largest increase in two years. Core inflation remains sticky, hovering around 7pc in urban areas and 8pc in rural areas. Additionally, energy price adjustments (including higher gas tariffs, subsidy removals, and increased fuel costs) and new tax measures are expected to feed into inflation in the near term. As a result, the SBP paused interest rate cuts in June and July, and the IIF expects interest rates to remain on hold for an extended period, with inflation averaging 6.5pc in FY26.
On the external front, the IIF forecasted that Pakistan’s current account will be influenced by the normalisation of imports (particularly machinery, raw materials, and consumer goods). Exports will depend largely on the progress of the US-Pakistan trade deal, though the IIF remains sceptical of its impact on exports. In the absence of a sharp increase in commodity prices, the current account deficit is expected to stay modest (about 0.5pc of GDP in FY26), allowing some reserve accumulation, though import coverage will remain critically low at about 2.5 months.
The IIF also expressed concerns over Pakistan’s fiscal situation. Although the country’s fiscal deficit narrowed to 5.4pc of GDP in FY25 and total revenues grew by 35.6pc, much of this growth came from one-off factors, such as record profits from the State Bank of Pakistan.
The Federal Board of Revenue (FBR) fell short of its target by about 1pc of GDP, and the federal tax-to-GDP ratio remains stuck at around 10pc. For FY26, authorities are forecasting another large increase in tax revenues, but this may be difficult to sustain given the already high tax burden on the formal sector and the exclusion of the retail/wholesale sectors, which account for about 20pc of GDP, from the tax net.
On the expenditure side, reductions in subsidies (particularly on electricity) and net lending to public enterprises have helped, but total spending still rose by 18pc in FY25. The IIF also noted that recent tensions with India could lead to an increase in defence spending in FY26.
As a result, the authorities’ targets of a 3.9pc of GDP deficit and a 2.4pc primary surplus for FY26 appear overly ambitious. The heavy reliance on domestic financing is also a cause for concern, and the IIF warned that fiscal performance will remain a key test for the IMF programme, with vested interests complicating the progress of reforms.
Establishing a mouse model of EPS with rapid intra-abdominal adhesions
We aimed to develop a novel and more efficient mouse model of encapsulating peritoneal sclerosis (EPS), closely mimicking the condition observed in human patients. To achieve this, we combined several high-risk factors commonly associated with peritoneal dialysis-related EPS, including a 4.25% peritoneal dialysis solution (PD), a surgical hygiene solution (SHS) consisting of chlorhexidine gluconate and ethanol, and lipopolysaccharide (LPS) to replicate episodic peritonitis (Fig. 1A). While traditional models using SHS alone require up to 8 weeks for EPS formation [13], the addition of PD and LPS successfully accelerated the process, enabling significant intra-abdominal adhesion formation by the 3rd week.
Fig. 1: Modifying an EPS mouse model to closely resemble PD-induced abdominal adhesions.
A Schematic of the experimental regimen for establishing the mouse model of EPS. B Gross macroscopic examination of the abdominal cavity. Control mice displayed smooth, well-expanded mesenteries and sharp liver edges; PD group and LPS group showed no observable changes; SHS-treated mice showed blunted liver edges (yellow arrow), but little interorgan adhesions. In the PD + LPS + SHS group, there were extensive adhesions between abdominal organs (blue arrow) and thickened liver margins caused by surface fibrosis. C Quantification of adhesion scores across Control group (n = 7), PD group (n = 6), LPS group (n = 7), SHS group (n = 16), and PD + SHS + LPS group (n = 25). D Dynamic body weight changes in mice over the course of the study. E Kaplan–Meier survival curve showing survival rates in each group. EPS mice obtained a survival rate of 83.3% (25/30) and a success rate of 100% (25/25). F Ultrasonographic comparison highlights the anatomical similarities between the murine EPS model and the human condition. a–d Ultrasound imaging of control mice revealed a smooth parietal peritoneum and normal gastrointestinal motility (a). In contrast, the EPS mice showed peritoneal thickening and calcification (b, arrows), adhesions between the parietal peritoneum and intestinal loops (c, arrows), and a characteristic “cauliflower-like” central clumping of small bowel loops (d). Supplementary Video 1 provides dynamic ultrasonography of the abdominal cavity in both control and EPS mice. e–h In human patients, control subjects demonstrated a smooth peritoneal surface (e), while EPS patients exhibited peritoneal thickening (f, arrows), calcification (g, arrows), and significant adhesions between the parietal peritoneum and intestinal loops (h, arrows). ***p < 0.001 by Student’s t test or ANOVA test. Fig. 1A, and the ultrasound, human, mouse drawing elements in Fig. 1F were created in BioRender. Sun, J. (2025) https://BioRender.com/vbajaol.
Macroscopic evaluation revealed distinct differences between the experimental groups. Control mice showed smooth hepatic edges and well-expanded mesenteries with no visible adhesions (Fig. 1B). In contrast, the PD + LPS + SHS group exhibited prominent signs of adhesion, including thickened liver margins with fibrous deposits (Fig. 1B, yellow arrow) and extensive adhesions between the abdominal organs, particularly between the liver, intestines, and parietal peritoneum (Fig. 1B, blue arrow). Additionally, these mice displayed severe mesentery contraction and intestinal dilation, with some cases forming an abdominal “cocoon” structure, indicative of advanced EPS pathology (Supplemental Fig. 1). The adhesion scores of the PD + LPS + SHS group were significantly higher than those of the single control groups, demonstrating the robustness of the model in replicating EPS (Fig. 1C).
To assess the physiological impact of EPS, we monitored body weight and survival rates. Mice in the PD + LPS + SHS group showed significant weight loss compared to other groups, suggesting a progression toward malnutrition or cachexia, a common complication in EPS (Fig. 1D). Despite the severity of the disease, the survival rate in this group remained relatively high at 83.3%, allowing us to reliably conduct further analyses (Fig. 1E).
Since invasive exploration of the abdominal cavity is impractical in clinical settings, we employed ultrasonography to non-invasively assess the structural changes in the peritoneum, as commonly done in patients20. Comparing to the control group (Fig. 1F-a, Supplementary Video 1), ultrasonography of EPS mice revealed key pathological features, including peritoneal thickening, calcification, and characteristic intestinal dilation (Figs. 1F-b and 1F-c), resembling the “concertina-like” appearance often observed in EPS patients. The mouse model also demonstrated strong adhesions between the parietal peritoneum and intestinal loops, which were confirmed by static and dynamic imaging (Fig. 1F-d, Supplementary Video 2). To validate the clinical relevance of our model, we compared these findings with ultrasonographic features from human EPS patients. The human subjects exhibited similar pathological traits, such as peritoneal thickening, calcification, and organ adhesions (Figs. 1F-f to 1F-h). These similarities between the mouse model and human disease underscore the utility of this model for studying EPS pathogenesis and evaluating potential therapeutic interventions. In summary, the combination of PD, LPS, and SHS successfully accelerated the development of EPS in mice, with key anatomical and pathological features closely mimicking human EPS. Given its efficiency and reproducibility, this model will be used for further studies aimed at understanding the mechanisms of EPS and testing potential therapeutic strategies.
EPS mouse model exhibits inflammation, fibrosis, and increased vascular density in peritoneum
To further characterize the pathological changes in our EPS mouse model, we performed detailed histopathological analyses of the peritoneum. As shown in Fig. 2A, cross-sectional images of the abdominal cavity revealed widespread and diffuse adhesions throughout the peritoneum in EPS mice. These adhesions formed dense, clot-like structures that encompassed multiple abdominal organs, closely resembling advanced EPS pathology observed in humans. Histological staining provided insight into the structural and cellular changes associated with EPS. Hematoxylin and eosin (H&E) staining indicated substantial thickening of both the parietal and visceral peritoneum in the EPS group, with marked infiltration of inflammatory cells (Fig. 2B, C). This infiltration signifies a heightened inflammatory response within the peritoneal tissues, which is a key feature of EPS progression. In addition, Masson’s trichrome staining highlighted extensive fibrotic deposition, further confirming that fibrosis is a central component of EPS pathology (Fig. 2B, C). Quantitative analysis of peritoneal thickness revealed a significant increase in EPS mice compared to controls, as depicted in Fig. 2G, H, reflecting the overall fibrotic burden in the disease. To explore the molecular drivers of this fibrotic response, we conducted immunohistochemical analyses, which demonstrated the upregulation of extracellular matrix (ECM) markers, including collagen type I alpha 1 (COL1A1), fibronectin (FN) and α-smooth muscle actin (α-SMA), in the peritoneum of EPS mice (Fig. 2D, Supplemental Fig. 2A). These markers are indicative of active fibroblast proliferation and matrix remodeling, key processes in tissue fibrosis.
Fig. 2: EPS mouse model exhibits fibrosis, inflammation, and increased vascular density in peritoneum, consisting with the pathological characteristics in human.
A Cross-sectional images of the abdominal cavity in EPS mice, showing widespread, diffuse adhesions forming clot-like structures throughout the peritoneal cavity. Hematoxylin and eosin (HE) and Masson’s trichrome staining of the parietal (B) and visceral (C) peritoneum, demonstrating significant peritoneal thickening and fibrotic deposition in EPS mice. D Immunohistochemical analysis of the parietal peritoneum showing increased expression of extracellular matrix (ECM) markers, including COL1A1, fibronectin (FN), and α-smooth muscle actin (α-SMA), indicative of active fibrosis. E Immunohistochemical analysis of inflammatory markers IL-1β, IL-6, and TNF-α in the parietal peritoneum, showing significant upregulation of these cytokines in EPS mice, highlighting the inflammatory component of the disease. F Immunofluorescent staining of CD31 in the visceral peritoneum (omentum) showing denser angiogenesis in EPS mice compared to controls. Quantification of peritoneal thickness in the parietal (G) and visceral (H) peritoneum, revealing significant thickening in EPS mice (n = 6 per group). I ELISA analysis of peritoneal lavage fluid, confirming an increase in IL-6 levels in EPS mice compared to controls (n = 6 per group). Data are presented as mean ± SEM.*** p < 0.001, two-tailed Student’s t-test.
Inflammation is known to play a critical role in EPS development21,22. In consistent with these observations, our immunohistochemical analysis showed significant upregulation of pro-inflammatory cytokines, including interleukin-1 beta (IL-1β), interleukin-6 (IL-6), and tumor necrosis factor-alpha (TNF-α), within the peritoneal tissues of EPS mice (Fig. 2E, Supplemental Fig. 2B). Furthermore, enzyme-linked immunosorbent assay (ELISA) confirmed elevated expression of IL-6 in the peritoneal lavage fluid of EPS mice compared to controls (Fig. 2I), highlighting the systemic inflammatory milieu. Another critical aspect of EPS pathology is increased vascular density, often associated with chronic inflammation and fibrosis. In the EPS model, we observed a notable increase in blood vessel density in the visceral peritoneum, as demonstrated by CD31 staining (Fig. 2F, Supplemental Fig. 2C). Collectively, these results provide a comprehensive view of the severe inflammation, fibrosis, and angiogenesis occurring in the peritoneum of EPS mice. Importantly, these pathological changes closely mirror those reported in human EPS, validating the utility of this mouse model for studying disease mechanisms and potential therapeutic interventions.
Dynamic collagen deposition and inflammatory cell infiltration during EPS formation
To investigate the dynamic progression of intraperitoneal adhesion formation in EPS, we sacrificed mice at different time points and analyzed the histopathological changes in the peritoneum. On day 1, no noticeable infiltration of inflammatory cells or collagen fiber deposition was observed on the peritoneal surface. By day 7, inflammatory cells had begun infiltrating the peritoneum, accompanied by the initial deposition of collagen fibers (Fig. 3A). This early infiltration laid the groundwork for adhesion formation, with immune cells accumulating on the surface of fibers by day 14, and the fibrils progressively thickening and interconnecting to form adhesions. As the process advanced, collagen fibers continued to proliferate and gradually became the dominant structural component of the adhesions, while the infiltration of immune cells began to subside. By day 21, the adhesions had evolved into dense fibrous scar tissue, with minimal remaining immune cell infiltration (Fig. 3A). These findings closely mirror the pathological evolution observed in human EPS, where early inflammation gives way to excessive fibrotic deposition over time.
Fig. 3: Histological features and dynamic formation of intraperitoneal adhesions in EPS mice.
A Representative images of H&E and Masson’s trichrome staining showing the dynamic pathological progression of adhesion formation in EPS mice. Mice were sacrificed at days 1, 7, 14, and 21 to capture different stages of adhesion development (AW: abdominal wall; BW: bowel wall). B Immunohistochemistry staining of adhesion cross-sections showing the distribution of key inflammatory cell types involved in adhesion formation. CD3 was used to identify lymphocytes, Ly6G for neutrophils, and F4/80 for macrophages. C Quantitative analysis of immune cell infiltrations in Fig. 3B (n = 6 per group). D A graphical diagram depicting the dynamic formation of interorgan adhesions, illustrating the transition from early inflammatory cell infiltration to fibrous scar tissue formation over time. Data are presented as mean ± SEM. *** p < 0.001, two-tailed Student’s t-test.
In addition to collagen depositions, H&E and Masson’s trichrome staining revealed significant infiltration of inflammatory cells within the adhesion regions (Fig. 3A). To determine the types of immune cells involved, we conducted immunohistochemical staining with markers for specific cell types at the conclusion of EPS molding (day 21). The markers used were F4/80 for macrophages, CD3 for T lymphocytes, and Ly6G for neutrophils (Fig. 3B). Among these, macrophages were found to be the predominant immune cells present within the adhesions, suggesting their key role in driving the fibrotic response (Fig. 3C). This was further supported by the extensive infiltration of macrophages in the parietal peritoneum (Supplemental Fig. 3), reinforcing the critical involvement of macrophages in the adhesion formation process. The graphical diagram in Fig. 3D illustrates the dynamic formation process of interorgan adhesions, depicting the transition from early inflammatory infiltration to the eventual deposition of fibrous scar tissue. The findings indicate the role of macrophage recruitment in the development of EPS, offering additional understanding of the cellular processes involved in adhesion formation within this disease model.
Transcriptomic alterations in peritoneal tissues of EPS mouse models
To gain insights into the transcriptomic alterations in the peritoneal tissues of EPS mice, we performed bulk RNA sequencing on the visceral peritoneum, using three biological replicates from both the Control and EPS groups (Fig. 4A). Our analysis identified a total of 3,900 upregulated genes and 4,397 downregulated genes in the EPS group compared to controls (adjusted p-value < 0.05). Notably, macrophage-associated genes such as Marco and Csf3r, along with chemokine-related genes, exhibited significant upregulation in the EPS group, underscoring the inflammatory nature of the disease (Supplemental Fig. 4).
Fig. 4: Transcriptomic analysis reveals increased fibrosis and inflammatory infiltration in the peritoneum of EPS mice.
A Heatmap of RNA-seq data comparing gene expression in the visceral peritoneum of Control (n = 3) and EPS (n = 3) mice. B Gene Ontology (GO) biological process enrichment analysis of differentially expressed genes (DEGs) significantly upregulated in EPS mice compared to controls. C GSEA) showing enhanced myeloid leukocyte activation and regulation of leukocyte adhesion in EPS mice. D UMAP plot illustrating cell clusters from two publicly available single-cell RNA sequencing (scRNA-seq) datasets of mouse omentum. Mes: mesothelial cells; Endo: Endothelial cells; Fib: Fibroblasts; Neutro: Neutrophils; Macro: Macrophages; T: T cells. E Venn diagram showing the overlap between cell type-specific marker genes from scRNA-seq datasets and DEGs identified in our bulk RNA-seq data. F Deconvolution analysis using scRNA-seq datasets to compare cellular proportions in the Control and EPS groups, highlighting an increased proportion of fibroblasts among parenchymal cells and macrophages among immune cells in EPS mice (n = 3 per group). G Flow cytometry analysis of visceral peritoneum from intestine, confirming a significant increase in immune cell infiltration, particularly macrophages, in EPS mice compared to controls (n = 5 per group). H Correlation analysis between COL1A1 expression and the infiltration of various inflammatory cell types, showing a strong positive correlation between macrophage infiltration and extracellular matrix (ECM) progression. I Correlation analysis between the macrophages proportion detected by flow cytometry with COL1A1 expression in the peritoneum by immunohistochemistry and peritoneal thickness. Data are presented as mean ± SEM. * p < 0.05, *** p < 0.001, two-tailed Student’s t-test.
To better understand the functional implications of these differentially expressed genes (DEGs), we performed Gene Ontology (GO) enrichment analysis. As shown in Fig. 4B, the upregulated genes in the EPS group were predominantly enriched in biological processes associated with immune activation, such as leukocyte migration, myeloid leukocyte activation, and interleukin production. These findings suggest that immune cell activation, particularly the infiltration of mononuclear macrophages, plays a central role in the progression of EPS. In addition, Gene Set Variation Analysis (GSVA) further supported this, with a marked increase in myeloid leukocyte activation scores in the EPS mice (Fig. 4C). This aligns with the observed increase in macrophage infiltration, indicating that these immune cells are key contributors to the ongoing inflammatory and fibrotic processes.
To gain further insight into the cellular composition of the peritoneal tissues, we utilized two publicly available single-cell RNA sequencing (scRNA-seq) datasets from mouse omentum and classified them into seven distinct clusters based on canonical marker genes (Fig. 4D) (GSE 136636, GSE176254). Differential expression analysis between the Control and EPS groups revealed that 63% of the marker genes identified from scRNA-seq overlapped with those from our bulk RNA-seq data (Fig. 4E). Using this dataset, we applied deconvolution analysis with the R package MuSiC to assess the relative proportions of different cell types. Our analysis showed a marked decrease in the proportion of mesothelial cells in the EPS group, suggesting a disruption in the mesothelial layer, which is a known contributor to adhesion formation by exposing adhesive fibrin clots to surrounding tissues23. Conversely, we observed a significant increase in the proportion of fibroblasts in the EPS mice, which likely contributes to the excessive fibrous deposition seen in the disease. In terms of immune cell populations, both macrophages and T cells were significantly elevated in the EPS group, with macrophages being the most predominant immune cell type (Fig. 4F). To validate these findings, we performed flow cytometry on peritoneal tissues, confirming that the percentage of macrophages was significantly higher in EPS mice, corroborating the RNA-seq results (Fig. 4G). Furthermore, we examined the correlation between different immune cell types and the expression of the fibrosis marker Col1a1. Among the immune cell markers evaluated, F4/80, a macrophage marker, demonstrated the strongest positive correlation with Col1a1 expression (R2 = 0.471, p = 0.042) (Fig. 4H). Concurrently, the proportion of macrophages detected by flow cytometry showed a significant positive correlation with COL1A1 expression in the peritoneum as measured by immunohistochemistry (R2 = 0.9311, p < 0.0001), as presented in Fig. 4I, suggesting that macrophages are the main immune cell type involved in fibrotic deposition in EPS. These findings, when combined with our previous immunohistochemistry data, emphasize the pivotal role of macrophage infiltration in the formation of EPS and further highlight the synergistic relationship between immune cell infiltration and fibrotic deposition in this disease model.
Activation of the cGAS-STING pathway in mesothelial cells regulates macrophage chemotaxis
To investigate the mechanism underlying the significant macrophage infiltration observed in the EPS peritoneum, we reanalyzed our bulk RNA sequencing data. KEGG analysis revealed that genes upregulated in EPS mice were enriched in pathways related to cell chemotaxis and cytoplasmic DNA sensing, particularly the cGAS–STING pathway (Supplemental Fig. 5A–B), highlighting a potential link between immune cell recruitment and intracellular immune surveillance mechanisms (Fig. 5A). Additionally, GSVA demonstrated a strong positive correlation between the cytoplasmic DNA-sensing pathway and cytokine chemotaxis and inflammatory response pathways, further supporting the role of intracellular surveillance system in EPS progression (Fig. 5B, C).
Fig. 5: Activation of the cGAS-STING pathway in mesothelial cells regulates macrophage chemotaxis.
A KEGG analysis showing significant enrichment of upregulated genes in cell chemotaxis and cytoplasmic DNA-sensing pathways in EPS mice. GSVA scores demonstrating a positive correlation between the cytoplasmic DNA-sensing pathway and cytokine chemotaxis (B) and inflammatory response pathways (C). D ELISA results showing elevated CCL2 levels in the peritoneal lavage fluid of EPS mice (n = 6 per group). E Immunofluorescence indicating increased CCL2 expression in the EPS parietal peritoneum, primarily co-localized with Cytokeratin 7+ mesothelial cells. F Immunofluorescence showing increased STING expression in the EPS parietal peritoneum, also co-localized with Cytokeratin 7+ mesothelial cells. G Western blot confirming the activation of cGAS-STING and downstream proteins in the EPS peritoneum (n = 6 per group). H Cytoplasmic DNA leakage observed in mesothelial cells stimulated with LPS + SHS. I, J Western blot and immunofluorescence showing STING activation and downstream signaling in mesothelial cells under EPS stimulation, with H151 reducing this activation (n = 4 independent experiments). K qPCR showing upregulation of inflammatory-related genes in mesothelial cells under EPS stimulation, with or without H151 pretreatment (n = 3 independent experiments). L qPCR demonstrating increased CCL2 gene expression in mesothelial cells under EPS stimulation, with H151 reducing the effect (n = 3 independent experiments). M Immunofluorescent staining of CCL2 in EPS-stimulated mesothelial cells. N ELISA showing elevated CCL2 levels in mesothelial cell supernatant under EPS stimulation, with H151 reducing CCL2 secretion (n = 3 independent experiments). O Schematic of the trans-well assay to assess macrophage migration. Mesothelial cells were seeded in the lower chamber, stimulated with EPS, and co-cultured with macrophages from the upper chamber (Created in BioRender. Sun, J. (2025) https://BioRender.com/vbajaol). P Bright-field image of macrophage migration after co-culturing with EPS-stimulated mesothelial cells. Q Quantification of macrophage migration under different conditions, with H151 alleviating macrophage migration (n = 3 independent experiments). R, S Representative image and quantification of macrophage migration under anti-CCL2 treatment, (n = 3 independent experiments). Data are presented as mean ± SEM. * p < 0.05, ** p < 0.01, *** p < 0.001, two-tailed Student’s t-test.
Given that CCL2 is a well-established chemokine involved in macrophage recruitment24,25, our qPCR validation confirmed that CCL2 exhibited the most pronounced expression change among chemokines in the peritoneal tissues of EPS mice (Supplemental Fig. 6A). ELISA analysis of the peritoneal lavage fluid revealed significantly higher levels of CCL2 in the EPS group compared to controls (Fig. 5D). Immunofluorescence also showed a marked increase in CCL2 expression in the parietal peritoneum, primarily co-localizing with Cytokeratin 7+ mesothelial cells (Fig. 5E, Supplemental Fig. 6B). These findings suggest that mesothelial cells are the primary source of CCL2 secretion in EPS, driving the chemotaxis of macrophages.
We then investigated the activation of the cGAS-STING pathway, which is known to mediate responses to cytoplasmic DNA. Both Western blot and immunofluorescence confirmed significant activation of the cGAS-STING pathway in EPS peritoneum, with STING co-localizing primarily with mesothelial cells (Fig. 5F, G, Supplement Supplemental Fig. 6C, D), suggesting that activation of the cGAS-STING pathway in mesothelial cells under EPS conditions may be responsible for promoting the chemotaxis of inflammatory cells, such as macrophages. After initial experiments to set up the in vitro EPS model (Supplemental Fig. 6E-H), we simulated the EPS environment by treating human peritoneal mesothelial cells (HPMCs) with LPS and SHS. Under these conditions, mesothelial cells exhibited cytoplasmic DNA leakage (Fig. 5H), consistent with cGAS-STING pathway activation. Western blotting and immunofluorescence confirmed the activation of STING and its downstream proteins in mesothelial cells upon EPS stimulation, while treatment with the STING inhibitor H151 partially mitigated this effect (Fig. 5I, J). qPCR analysis further demonstrated that the expression of inflammatory-related genes was significantly upregulated in mesothelial cells under EPS stimulation, and this effect was also suppressed by H151 (Fig. 5K). We next examined the role of CCL2 in this process. RT-qPCR and cellular immunofluorescence showed that both CCL2 gene expression and CCL2 protein levels were significantly elevated in mesothelial cells under EPS stimulation, and these increases were attenuated by H151 (Fig. 5L, M). The secretion of CCL2 in the cell supernatant was also significantly higher in EPS-stimulated mesothelial cells (Fig. 5N). STING agonists ADU-S100 also significantly induced CCL2 expression (Supplemental Fig. 6I, J), verifying the role of STING activation in promoting CCL2 secretion.
To directly assess the impact of mesothelial cell activation on macrophage migration, we performed a transwell co-culture assay, where mesothelial cells were co-cultured with macrophages (Fig. 5O). EPS-stimulated mesothelial cells significantly promoted macrophage migration (Fig. 5P, Q), while treatment with H151 partially alleviated this effect. These findings demonstrate that activation of the cGAS-STING pathway in mesothelial cells leads to the secretion of CCL2, which in turn induces macrophage chemotaxis and migration. Collectively, our results highlight the critical role of mesothelial cells in sensing EPS-related stimuli and activating the cGAS-STING pathway, thereby driving macrophage recruitment through CCL2 secretion. Inhibition of STING effectively reduces both CCL2 production and macrophage migration, offering a potential therapeutic approach for mitigating macrophage-driven inflammation in EPS. Additionally, we observed that anti-CCL2 intervention also significantly attenuated in vitro EPS-induced macrophage migration, again indicating that CCL2 acts as the predominant cytokine driving macrophage chemotaxis in this microenvironment (Fig. 5R, S).
Inhibition of cGAS-STING activation ameliorates EPS formation in mice
To investigate whether inhibiting the cGAS-STING pathway could reduce the severity of EPS, we administered the STING inhibitor H151 via intraperitoneal injection in mice (Fig. 6A). Western blot and immunofluorescence analyses confirmed that H151 effectively suppressed the activation of STING and downstream signaling proteins in the peritoneal tissues (Fig. 6B), validating its ability to block STING signaling in vivo. In parallel, H151 treatment significantly lowered the expression of CCL2 in peritoneal tissue (Fig. 6C, Supplemental Fig. 7A) and in the peritoneal lavage fluid (Fig. 6D). This reduction in CCL2 was associated with a decrease in macrophage infiltration on the peritoneal surface, observed through immunohistochemistry (Fig. 6E, F). Correlation analysis between the CCL2 concentration in peritoneal lavage fluid and the F4/80 infiltration area showed a significant positive correlation (Fig. 6G). These results indicate that STING-mediated chemotaxis, driven by CCL2, plays a central role in recruiting macrophages during EPS progression. Therapeutically, H151 treatment markedly improved the pathological features of EPS. Mice treated with H151 showed a significant reduction in adhesion scores during gross assessments (Fig. 6H, I), reflecting decreased intra-abdominal adhesions. Histological staining also revealed a notable reduction in collagen fiber deposition in the adhesion regions (Fig. 6Ja-b and 6K). The thickness of the parietal peritoneum, a hallmark of fibrosis in EPS, was significantly decreased in H151-treated mice (Fig. 6J-c and L), indicating a direct impact on fibrotic progression. Immunohistochemistry showed that ECM-related proteins (COL1A1, fibronectin, α-SMA) (Fig. 6d-f), inflammatory markers (IL-1β, IL-6, TNF-α), and pro-angiogenic VEGF were significantly reduced in the H151-treated group (Fig. 6M), supporting the inhibitor’s anti-adhesion effect. In summary, these findings suggest that inhibiting cGAS-STING activation with H151 reduces CCL2 secretion and macrophage recruitment, ultimately mitigating the excessive collagen deposition and adhesion formation characteristic of EPS. This highlights the potential of targeting the STING pathway as a therapeutic strategy for EPS.
Fig. 6: Inhibition of the cGAS-STING activation effectively ameliorated abdominal adhesion in EPS.
A The schematic diagram of administering the STING inhibitor H151 (Created in BioRender. Sun, J. (2025) https://BioRender.com/vbajaol). Western blot (B) and immunofluorescence (C) demonstrating that H151 effectively reduced the activation of STING and its downstream proteins in peritoneal tissues (n = 6 per group). D ELISA exhibited that H151 partially lowered the CCL2 concentration in peritoneal lavage fluids (n = 6 per group). E–F Immunohistochemistry (IHC) showed H151 effectively reduced macrophage infiltration (n = 6 per group). G Correlation analysis between the CCL2 concentration in peritoneal lavage fluid and the F4/80 infiltration from (F) among the 3 groups. Gross macroscopic viewing of abdomen (H) and the macroscopic adhesion score (I) demonstrated from a macro perspective that H151 effectively alleviated abdominal adhesion (n = 6 per group). J–L Histopathological assessment of intra-abdominal adhesions and fibrous deposition on peritoneal surface. a MASSON staining of the abdominal cross-sections presented the adhesion condition in different groups, and the MASSON+ area (b) was calculated and the statistic diagram is shown in Figure K (n = 6 per group). c Masson’s trichrome staining illustrates the thickness of the parietal peritoneum among the three groups, with (L) showing the corresponding statistical bar graph (n = 6 per group). d–f Immunohistochemical images presenting the expression of three classic ECM-related proteins in parietal peritoneum: COL1A1, Fibronectin, and α-SMA. M Immunohistochemical analysis of inflammatory markers IL-1β, IL-6, and TNF-α in the parietal peritoneum. Data are presented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, two-tailed Student’s t-test.
IRF3 as the key transcription factor promoting CCL2 secretion in mesothelial cells
Cyclic GMP-AMP synthase (cGAS) detects cytosolic DNA during cellular stress and activates the adaptor protein STING, which triggers immune responses by activating the downstream transcription factor IRF326,27,28. We investigated whether the activation of STING in mesothelial cells under EPS condition regulates CCL2 secretion via IRF3. Using the Univariate Linear Model (ULM) method in the DecoupleR package, we analyzed our bulk RNA-seq data and found that IRF3 transcriptional activity was significantly elevated in the EPS group (Fig. 7A). This increase in IRF3 activity aligns with our observation of increased phospho-IRF3 levels in mesothelial cells under EPS condition (Fig. 5I). Among the genes predicted to be regulated by IRF3, CCL2 and CCL5 were the most upregulated in the EPS group (Fig. 7B), with CCL2 had the higher p-value (p < 10-6). To further investigate how IRF3 promotes CCl2 expression, we used the JASPAR database to analyze the transcription factor binding sites within the CCL2 promoter and identified a high-scoring potential IRF3 binding site (Fig. 7C). This suggested a direct interaction between IRF3 and the CCL2 promoter region (Fig. 7D). To test this possibility, we conducted a chromatin immunoprecipitation (ChIP) assay, which validated that IRF3 indeed binds to the promoter of the CCL2 gene (Fig. 7E, F, Supplemental Fig. 8). These findings demonstrate that under EPS condition, IRF3 binding to the CCL2 promoter facilitates increased CCL2 production and secretion. Thus, IRF3 serves as a key transcription factor driving CCL2 secretion in mesothelial cells in response to EPS. IRF3 knockdown significantly reduced both the transcription and secretion of CCL2 in response to stimulation with a STING agonist (Supplemental Fig. 6I, J) and EPS-conditioned media (Fig. 7G, H). Furthermore, this genetic intervention markedly inhibited macrophage migration, as shown by transwell assays (Fig. 7I, J). These results provide functional evidence that IRF3 is a key transcriptional regulator of CCL2 in EPS-associated mesothelial responses.
Fig. 7: IRF3 as the key transcription factor promoting CCL2 secretion in mesothelial cells.
A Bar plot showing the activity scores of differentially expressed transcription factors in Saline and EPS group, based on bulk RNA-seq data. B Volcano plot displaying target genes of IRF3 that are differentially expressed between Saline and EPS group. Red dots indicate genes activated by IRF3, while blue dots represent genes inhibited by IRF3. C Predicted IRF3 binding motifs identified using the JASPAR database. D Schematic illustration showing the IRF3 binding motif within the promoter region of the CCL2 gene. E–F ChIP assays and ChIP-qPCR analysis of IRF3 binding to CCL2 in mesothelial cells treated with or without EPS (SHS + LPS) and H151 (n = 3 independent experiments). Gene silencing of IRF3 significantly reduced both the transcriptional expression (G) and secretion (H) of CCL2 induced by SHS + LPS. I–J Trans-well experiments demonstrating IRF3 silencing suppressed EPS-induced macrophage migration. Data are presented as mean ± SEM. ***p < 0.001, two-tailed Student’s t-test.
Clinical relevance of STING activation and EPS
For a clinical investigation of the cGAS-STING pathway, we collected peritoneal tissues, including adhesion tissues, from patients with EPS and healthy controls. Immunofluorescence staining revealed significant activation of the cGAS-STING pathway in mesothelial cells from the peritoneal surface of EPS patients (Fig. 8A), along with a marked increase in CCL2 expression (Fig. 8B). Given that a significant number of mesothelial cells are shed into the peritoneal dialysis fluid under EPS condition, we also prepared smears of the shed cells from the peritoneal dialysis effluent for immunofluorescence staining, which yielded results consistent with the peritoneal tissue (Fig. 8C–H). Additionally, ELISA measurements confirmed that the concentration of CCL2 in the peritoneal dialysis effluent was significantly higher in EPS patients compared to controls (Fig. 8I), so as the cGAMP, which serves as an indicator of STING activation (Fig. 8J). Furthermore, there is a positive correlation between CCL2 and cGAMP concentrations (Fig. 8K). These clinical findings corroborate the link between STING activation and increased CCL2 levels in EPS, further supporting the role of the cGAS-STING pathway in driving macrophage chemotaxis and inflammation.
Fig. 8: STING and CCL2 are upregulated in parietal peritoneum of EPS patients.
HE staining and immunofluorescence of peritoneal tissue from clinical patients showed the increased expression of STING (A) and CCL2 (B) in EPS patients compared to healthy control (n = 6 per group). C–H Immunofluorescence of peritoneal dialysis effluent smear showed the increased expression of CCL2 (C), STING (D), and p-IRF3 (E) in EPS patients compared to short-term PD patients (SPD) (n = 6 per group). With the corresponding quantitative statistics of fluorescence co-localization showing in F, G and H. I–J ELISA measurements revealed a significant increased CCL2 concentrations (E) and cGAMP concentrations (F) in the peritoneal dialysis effluent in EPS patients compared to control (n = 6 per group). K Correlation analysis presented a significant positive correlation between cGAMP and CCL2 (n = 12). Data are presented as mean ± SEM. *p < 0.05, ***p < 0.001, two-tailed Student’s t-test.
The State Bank of Pakistan (SBP) injected a staggering Rs1.4 trillion into the financial system on Friday through dual open market operations, using both conventional and Shariah-compliant instruments to manage the huge liquidity demand.
In the conventional reverse repo operation, the SBP injected Rs1.037 trillion against total offers of Rs1.150 trillion. The central bank accepted Rs76 billion at 11.06% for seven-day tenor and Rs962 billion at a cut-off rate of 11.04% for 13-day tenor, with pro rata allotment applied where necessary.
Simultaneously, the SBP conducted a Shariah-compliant Mudarabah-based injection of Rs363 billion, accepting all offered bids – Rs241 billion at 11.14% for seven days and Rs122 billion at 11.13% for 13 days.
Moreover, the Pakistani rupee registered a marginal gain against the US dollar, appreciating 0.01% in the inter-bank market. At close, the local currency settled at 281.90, up two paisa compared with the previous day’s close at 281.92. This marked the rupee’s 11th consecutive session of gains against the greenback.
According to Ismail Iqbal Securities, the rupee has now appreciated 0.66% in the current fiscal year to date, although it remains down 1.19% on a calendar-year-to-date basis.
Analysts at AKD Securities highlighted that the rupee has strengthened for the fifth consecutive week, reflecting improved sentiment amid stability in foreign exchange reserves and remittance inflows.
Meanwhile, gold prices in Pakistan fell, contrary to movements in the international market, where bullion rebounded after comments from US Federal Reserve Chair Jerome Powell fueled expectations of a September rate cut at the Jackson Hole symposium.
According to the All Pakistan Sarafa Gems and Jewellers Association, the price of gold declined Rs1,500 to settle at Rs355,700 per tola, while the rate for 10 grams dropped Rs1,286 to Rs304,955. A day earlier, gold had gained Rs2,000 to close at Rs357,200 per tola.
Internationally, spot gold was up 0.7% at $3,362.53 per ounce by 10:26 am EDT (1426 GMT), while US gold futures were 0.8% lower at $3,408.20, Reuters reported.
Market analysts noted that initially gold traded in a narrow $25-40 range, with little momentum amid lack of progress in the geopolitical situation such as the Russia-Ukraine conflict. However, Powell’s remarks that future data could warrant interest rate cuts triggered renewed buying, lifting prices to the high of $3,380.
Interactive Commodities Director Adnan Agar said that the market remained range bound but faced strong resistance at the $3,400 level. “If gold breaks this barrier, the next target is projected around $3,450,” he said.
Agar added that expectations of a downward shift in US interest rates, combined with persistent geopolitical risks, continue to bolster the metal’s appeal as a safe-haven asset.
With US rates having remained unchanged for an extended period, the possibility of an imminent cut has reinforced bullish sentiment, though traders caution that volatility will likely persist in the short term.
Coordinating laboratory: Teagasc Food Research Centre, Moorepark, Fermoy, Co Cork P61 C996, Ireland.
Participating laboratories:
Laboratory of Food Chemistry and Biochemistry, Department of Food Science and Technology, School of Agriculture, Aristotle University of Thessaloniki, P.O. Box 235, 54124, Thessaloniki, Greece
Global Oatly Science and Innovation Centre, Rydbergs Torg 11, Space Building, Science Village, 22 484 Lund, Sweden
Laboratory of Food Technology, Department of Microbial and Molecular Systems (M2S), KU Leuven, Kasteelpark Arenberg 23, PB 2457, 3001, Leuven, Belgium
INRAE, Institut Agro, STLO, 35042 Rennes, France
School of Biosciences, Faculty of Health and Medical Sciences, University of Surrey, Guildford, GU2 7XH, United Kingdom
Nofima AS, Norwegian Institute of Food, Fisheries and Aquaculture Research, PB 210, N-1433, Ås, Norway
Center for Innovative Food (CiFOOD), Department of Food Science, Aarhus University, Agro Food Park 48, Aarhus N 8200, Denmark
Department of Horticulture, Martin-Gatton College of Agriculture, Food and Environment, University of Kentucky, Lexington, Kentucky, USA
Department of Agricultural, Food, Environmental and Animal Sciences, University of Udine, Italy
Wageningen Food & Biobased Research, Wageningen University & Research, 6708 WG Wageningen, The Netherlands
Quadram Institute Bioscience, Rosalind Franklin Road, Norwich Research Park, Norwich, NR4 7UQ, United Kingdom
Department of Food Engineering, Faculty of Engineering, Ege University, 35100, İzmir, Türkiye
Materials
The chemicals and four test products used in the ring study are presented in (Table 3). They were ordered by the coordinating laboratory, aliquoted and shipped to each of the participating laboratories. All laboratories received aliquots from the same batch of each product, with the exception of 3,5-dinitrosalicylic acid (DNSA) which came from two different lots. Prior to shipping, calibration curves established with solutions prepared from both of these lots were compared, and showed nearly equivalent results (Figure S1 in Supplementary material-Section “Protocol implementation at each laboratory”).
Table 3 Products supplied to the laboratories participating in the ring trial.
Equipment needed
The list of equipment required is provided as guidance below.
Preparation of reagents and enzyme solutions
Vortex mixer, pH meter with glass electrode, heating/stirring plate, incubator.
Enzyme assay
Water-bath or thermal shaker (e.g. PCMT Thermoshaker, Grant Instruments, United Kingdom) for enzyme–substrate incubations at 37 °C. Boiling bath (e.g. SBB Aqua 5 Plus, Grant Instruments, United Kingdom) or thermal shaker (e.g. PCMT Thermoshaker, Grant Instruments, United Kingdom) suitable for use at 100 °C. Spectrophotometer (e.g. Shimadzu UV-1800 Spectrophotometer, Shimadzu Corporation, Japan) or plate reader (e.g. BMG Labtech CLARIOstar Plus, BMG Labtech, Germany).
Basic materials
Volumetric flasks, heatproof bottle with lid (e.g. Duran bottle), magnetic stirrer, timer, thermocouple, safe lock microtubes (2 or 1.5 mL), heat (and water) resistant pen or labels for the microtubes, disposable standard cuvette or disposable polystyrene 96-well plate.
Preparation of reagents and enzymes
20 mM Sodium phosphate buffer (with 6.7 mM sodium chloride, pH 6.9 ± 0.3)
Prepare a stock solution by dissolving 1.22 g NaH2PO4 (anhydrous form), 1.38 g Na2HPO4 (anhydrous form) and 0.39 g NaCl in 90 mL purified water and make up the volume to 100 mL. Before use, dilute 10 mL of stock solution to 95 mL with purified water. Confirm that the pH of the buffer, when heated to 37 °C, is within the required working range (pH 6.9 ± 0.3). If needed, adjust the pH, using 1 M NaOH or HCl as required, before making up the volume to 100 mL.
Maltose calibrators
Prepare a 2% (w/v) maltose stock solution in phosphate buffer. Prepare a calibrator series by diluting the maltose stock solution in phosphate buffer as indicated in Table S2 (Supplementary Material – Section “Protocol implementation at each laboratory”). Store in the fridge (or freezer if not for use during the same day).
Colour reagent (96 mM DNSA with 1.06 M sodium potassium tartrate)
Dissolve 1.10 g of DNSA in 80 mL of 0.50 M NaOH at 70 °C in a glass beaker or bottle (partly covered to limit evaporation) on a pre-heated heat/stir plate with continuous stirring and temperature monitoring (e.g. using a thermocouple). Once the DNSA is fully dissolved, add 30 g of sodium potassium tartrate and continue stirring until it dissolves. Remove from heat and wait until the solution cools to room temperature. Bring to 100 mL with purified water. Store at room temperature protected from light for up to 6 months. If precipitation occurs during storage, re-heat to 45 °C while stirring on a heat-stir plate.
Starch solution
Potato starch pre-gelatinized in sodium phosphate buffer (1.0% w/v) is used as substrate. Pre-heat a heat-stir plate (setting it to 250 °C—300 °C is suggested) and pre-heat an incubator (or water bath) to 37 °C. Weigh 250 mg of potato starch into a heatproof bottle and add 750 μL of ethanol (80% v/v). Stir on a vortex mixer to wet all the starch powder (this is a critical step for the complete solubilisation of the starch). Add 20 mL of sodium phosphate buffer and mix again using a vortex mixer making sure that the powder is fully dispersed and there are no lumps in the solution. Cover the bottle with the lid to minimize evaporation (but making sure it is loose enough to let out excess steam) and place on the pre-heated heat-stir plate stirring at 180 rpm. When the solution starts bubbling, start the timer and boil on the heat-stir plate stirring continuously for exactly 15 min. Cool in the incubator/water bath for 15 min (or until it is safe to handle). Make up the volume of the starch solution to 25 mL in a volumetric flask by adding purified water. Store the solution in a closed bottle in an incubator (or water bath set to 37 °C) and use within 2 h. If the starch solution does not clarify significantly a new solution needs to be prepared, as this may indicate poor solubilisation and or gelatinization of the starch. Prepare a fresh solution each time as storing or freezing can cause starch retrogradation and influence the results of the assay.
α-amylase solutions
The preparation of the enzyme solutions is a critical step. Solutions prepared from enzyme powders should be carefully prepared following the same protocol each time to ensure adequate powder hydration and dispersion. After weighing the enzyme powder and adding the adequate amount of sodium phosphate buffer, stock solutions should be stirred in an ice bath (at around 250 rpm) for 20 min before any further dilutions (Graphical protocol in Fig. 6 and Picture S1 in the Supplementary Material). Subsequent dilution(s) of the stock solution(s) should be performed using sodium phosphate buffer to reach the recommended enzyme concentration of 1.0 ± 0.2 U/mL. For the four products tested in the ring trial, recommended concentrations are provided as reference in Table S7 (Supplementary material). For enzyme preparations, it is recommended to start from a stock solution prepared by adding 20 – 100 mg of enzyme powder to 25 mL of sodium phosphate buffer. For human saliva, a stock solution can be prepared by mixing 80 µL of saliva with 920 µL of buffer.
Fig. 6
Schematic overview of the enzyme assay. Created in BioRender.com.
Each enzyme should be tested at three different concentrations prepared by diluting 0.65 mL, 1.00 and 1.50 mL of enzyme stock solution with 1.35, 1.00 and 0.50 mL of buffer, respectively (Table S3). These diluted enzyme solutions are referred to as solutions C1, C2 and C3. Enzyme solutions should always be kept on ice and used within 30 min of preparation.
Enzymatic assay
An overview of the enzyme assay is presented in (Fig. 6).
Preparative procedures
Before starting, the following preparations are recommended: set the heating-block (water bath) as required to ensure 37 °C inside the microtubes (see troubleshooting advice, Table 2); pre-warm the starch solution to 37 °C; prepare a polystyrene container with ice.
Sample collection tubes
For each incubation that will be carried out, label and pre-fill four microtubes with 75 μL of DNSA colour reagent.
Incubations
Set three microtubes (one for each diluted enzyme solution C1, C2 and C3) in the preheated thermal shaker and let the temperature equilibrate before adding 500 µL of pre-warmed potato starch solution to each tube (maintain the tubes closed until the enzyme is added to prevent evaporation). Add 500 µL of diluted enzyme solution C1, C2 and C3 to the corresponding tubes at regular intervals. It is recommended to start the timer immediately when the α-amylase solution is added to the first tube and leave a 30 s interval before each subsequent addition.
Sample collection
Take a 150 μL aliquot from each tube after 3, 6, 9 and 12 min of incubation (respecting the order and intervals at which the incubations were initiated) and transfer it immediately to the corresponding sample collection tube pre-filled with DNSA to stop the reaction. Each aliquot should be taken as closely as possible to its respective sample collection time, within a maximum of ± 5 s.
Absorbance measurements
Prepare the maltose calibrators by mixing 150 µL of each maltose calibrator with 75 µL of DNSA reagent. Centrifuge the samples and calibrators (1000 g, 2 min) so that all droplets are brought back into solution. Place the samples and calibrators in the thermal shaker (or boiling bath) (100 °C, 15 min) and then transfer them to an icebox to cool for 15 min. Add 675 µL of purified water to each tube and mix by inversion. Transfer the samples and calibrators to a cuvette or pipette to a microtiter plate (300 µL per well) and record the absorbance at 540 nm (A540nm).
Ring trial organization
Preliminary testing
Throughout the protocol optimization phase, the assay was repeated multiple times by the coordinating laboratory to define practical aspects. Each of the four test products has been assayed at different concentrations. The final test concentrations were defined by choosing a test concentration that allowed for an adequate distribution of the endpoint measure (spectrophotometry absorbance) and communicated to the participating laboratories.
Protocol transference
A detailed written protocol (Supplementary material) was transferred to each participating laboratory including the recommendations for concentrations of the test products. All laboratories were invited to an online training session that included a video of the assay followed by a Q&A session to clarify any doubts. All labs carried out the assay and reported their results on a standard Excel file between May and November 2023.
Incubation temperatures
All laboratories tested the four enzyme preparations at 37 °C as described above. A subgroup of five laboratories also repeated the assays at 20 °C with the purpose of trying to establish a correlation between the results obtained at both temperatures.
For incubations at 20 °C protocol adaptations were performed as follows. A different recipe was used to prepare the 200 mM sodium phosphate buffer stock solution. It consisted of 1.26 g NaH2PO4, 1.29 g Na2HPO4 and 0.39 g NaCl. The dilutions (10 mL stock diluted to 95 mL with purified water) and pH (6.9 ± 0.3) were the same as those for the buffer used at 37 °C. All reagents and solutions requiring the use of buffer were freshly prepared using this buffer recipe. The recommended concentrations of the α-amylase stock solutions were adjusted to ensure that enough enzymatic activity was present.
Calculations
Calibration curve
The A540nm of the colour reagent blank was subtracted from the readings of all maltose calibrators and their concentration (mg/mL) was plotted against the corresponding ΔA540nm. For reference purposes, using a 96 well plate, the absorbance at 540 nm should increase linearly from approximately 0.05 (for the colour reagent blank) to 1.5 for the highest maltose concentration. The calibration blank should not be included as a data point in the calibration curve.
Enzyme activity definition
The definition of α-amylase activity resulting from the application of the newly developed protocol is the following:
Based on the definition originally proposed by Bernfeld: one unit liberates 1.0 mg of maltose equivalents from potato starch in 3 min at pH 6.9 at 37 °C.
Based on the international enzyme unit definition standards: one unit liberates 1.0 μmol of maltose equivalents from potato starch in 1 min at pH 6.9 at 37 °C.
Amylase activity units based on the definition originally proposed by Bernfeld were multiplied by the conversion factor 0.97 to convert the result into IU.
Enzyme activity calculation
The first step was to subtract A540nm of the colour reagent blank from all readings. The calibration curve was then used to calculate the maltose concentrations (mg/mL) reached with each diluted enzyme solution (C1, C2 and C3) at each sampling point during incubations. Enzyme concentrations during incubations were then calculated as mg/mL for enzyme powders, or µL/mL for liquid (saliva) samples.
For each diluted enzyme solution (C1, C2 or C3), maltose concentrations (mg/mL) were plotted against time (tmin) and the corresponding linear regression was established to determine the reaction kinetics’ slope ((text{m}t{text{min}})). For each enzyme concentration, units of enzyme were calculated using the following equation.
$$Activity (U per mg or mu L of enzyme product)= 3mintimes frac{text{m}t{text{min}}(frac{maltose concentration (frac{mg}{mL})}{time (min)})}{Enzyme concentration left(frac{mg}{mL} orfrac{mu L}{mL}right)}$$
A template Excel file is provided for calculations in the Supplementary Material.
Statistical analysis and assessment of method’s performance
Data visualization and statistical analyses have been performed in R (version 4.3.2)29. The packages ggplot230 and ggdist31 have been used in the preparation of the plots presented in the manuscript.
Outlier analysis was conducted on non-transformed data to preserve the original variability and scale of the datasets. First, Cochran’s test (outliers package in R32) was used to assess intralaboratory variability and did not reveal any outliers. Subsequently, for interlaboratory comparisons, boxplot analysis, Bias Z-scores and Grubbs’ test32 were employed complementarily. The results reported by one lab for three test products (pancreatin, α-amylase M and α-amylase S) assayed at 37 °C were more than 1.5 interquartile ranges below the 25th or above the 75th percentiles, consistent with unsatisfactory Bias Z-scores (|z|> 3). Grubb’s test confirmed these as outliers and they have been excluded from the statistical analysis. All results in the 20 °C dataset fell within 1.5 interquartile ranges of the 25th and 75th percentiles (Fig. 5), consistent with satisfactory Bias Z-scores (|z|< 2) (Supplementary Figure S4). While Grubbs’ test identified two potential outliers (Lab A for pancreatin and Lab D for α-amylase M), this outcome was considered less reliable due to the small sample size (n = 5) and lack of corroboration from boxplot and Bias Z-score analyses, and so these results were retained.
Statistical analysis of the dataset resulting from the implementation of the protocol at 37 °C has been carried out to investigate the effects of the tested products, concentrations and incubation conditions (thermal shaker vs. water bath with or without shaking) as well as the two-way and three-way interactions between these factors. Normality of this dataset has been confirmed through the Shapiro–Wilk test (p > 0.05). The homogeneity of variances, as assessed using Levene’s test in the Rstatix package version 0.7.233, was not confirmed (p < 0.001). Due to the limited availability of suitable non-parametric alternatives, a logarithm transformation was performed on this data set enabling homogenisation of the variances and application of a three-way ANOVA (Rstatix package). Statistically significant effects were further examined using Pairwise T-Test comparisons, applying Bonferroni adjustments for multiple comparisons as required. The results obtained when implementing the protocol at 20 °C were normally distributed, but homogeneity of variances was not confirmed for this dataset either. The corresponding logarithm transformed data frame did not conform to normality, hence the Kruskal–Wallis test was applied to examine the significance of the differences between the four products, followed by the Bonferroni-corrected Wilcoxon test for pairwise comparisons (all tests performed using the Rstatix package). Statistically significant effects have been accepted at the 95% level.
For each laboratory and product, an individual ratio of α-amylase activity at 37 °C to 20 °C was calculated, and the mean of these ratios across all laboratories was determined for each product. The 95% confidence interval for this mean ratio was computed using the t-distribution. Normal distribution and homogeneity of variances have been confirmed for this dataset, hence one-way ANOVA was used to investigate whether the ratios obtained for each product were significantly different.
For a thorough understanding of the method’s reliability, precision, and transferability across different laboratory settings three complementary metrics have been used: Z-scores based on bias scores for a standardized evaluation of systematic errors, repeatability and reproducibility.
Z-scores were calculated to standardize the comparison of bias scores across laboratories and products enabling to assess the overall agreement between individual laboratory results and the mean for each product. For each product, bias scores were first calculated for each laboratory using the mean of all laboratories as the reference value and then converted to z-scores:
$$text{z }=frac{left( x -text{ X}right)}{text{SD}}$$
x is the individual laboratory result, X is the mean of all laboratories, and SD is the standard deviation. Z-scores interpretation followed standard criteria with |z|≤ 2 as satisfactory, 2 <|z|< 3 as questionable, and |z|≥ 3 as potentially unsatisfactory.
Repeatability (measured as intralaboratory coefficient of variation, CVr), which quantifies method precision within each laboratory, reflecting consistency under identical conditions, was calculated as the root mean square of the individual laboratory’s CVs:
CVr is the coefficient of variation under repeatability conditions (intralaboratory); (i) indexes each laboratory, ({CV}_{i}) is the coefficient of variation for laboratory (i); L is the number of participating laboratories.
Reproducibility (measured as coefficient of variation, CVR), a measurement of method’s consistency across different laboratories indicates its robustness to varying environments and operators, was calculated for each tested product as:
$${CV}_{R}=frac{SD}{X} times 100$$
CVR is the coefficient of variation under reproducibility conditions (interlaboratory); SD and X correspond to the standard deviation and mean values calculated from interlaboratory data.
Coefficients of variation below 30%15,16 are frequently considered to be indicators of small intra- and interlaboratory variability. In some cases, critical thresholds for repeatability (intralaboratory CV) are set at 20%34.
