Identifying Immune-related Molecular Biomarkers in Autism Spectrum Disorder Using Data-independent Acquisition Proteomics and Machine Learning.

Scientists studied blood samples from 99 children with autism and 70 children without autism to find protein markers that might help identify autism. They used advanced laboratory techniques and computer analysis to find eight immune system proteins that were different between the two groups. A computer model using these proteins was 95% accurate at telling the difference between children with and without autism.

This study developed a protocol for identifying autism spectrum disorder (ASD) biomarkers using advanced protein analysis combined with machine learning. Researchers analyzed blood samples from 99 children with ASD and 70 controls using data-independent acquisition mass spectrometry to profile proteins in the blood. Machine learning algorithms identified eight immune-related proteins that could distinguish between ASD and control groups. A predictive model based on these proteins achieved 95.27% accuracy in differentiating ASD cases from controls.

The study demonstrates the potential of combining advanced protein analysis techniques with machine learning for biomarker discovery in autism research.

While showing promising accuracy, these biomarkers require validation in larger, independent populations before clinical application. The immune-related protein findings align with existing autism-immunity research but need replication studies to establish clinical utility.

Published in a methods journal suggesting focus on protocol development rather than clinical validation. Cross-validation results may not reflect real-world performance. No information provided about replication in independent cohorts or clinical validation studies.

This study presents a reproducible protocol for identifying serum protein biomarkers associated with autism spectrum disorder (ASD) using data-independent acquisition (DIA) mass spectrometry combined with machine learning (ML). DIA enables unbiased, high-resolution profiling of the serum proteome, including low-abundance proteins, while ensuring reproducibility across samples. ML approaches were applied to select diagnostically informative protein panels and improve model robustness. The analysis included serum from 99 children with ASD and 70 age-matched controls.

High-abundance proteins were depleted, peptides were prepared using standardized digestion and fractionation procedures, and DIA was performed on a high-resolution mass spectrometer. Data processing and quantification identified differentially expressed proteins, which underwent functional enrichment analysis. Eight immune-related proteins emerged as strong candidates for biomarker development. A logistic regression model trained on these proteins achieved 95.27% accuracy, a Kappa value of 0.9025, and an AUC of 1.000 in cross-validation.

These findings demonstrate the potential of DIA-based proteomics, combined with machine learning, as a robust framework for biomarker discovery in ASD and for adaptation in broader clinical research.