Biomarker Validation & Trial Design Consulting | Translational Multi-Omics | Neurodegeneration & Metabolic Disease | Systems Biology | Behavioral Science | Machine Learning & A.I.

Independent consultant specializing in biomarker validation and clinical trial design for neurodegenerative disease, metabolic disease, and digital health programs. I focus on preventing costly failures by identifying confounding vulnerabilities and methodological flaws before trials enroll or studies publish.

15+ years: Multi-omic integration, translational neuroscience, clinical trial and observational study design, deep expertise in longitudinal studies

Former: Senior Research Scientist, Institute for Systems Biology (ISB; worked with Lee Hood and Mary Brunkow); Assistant Professor, University of Houston

Editor: Academia Neuroscience & Brain Research

Track record: 65+ publications, $5M+ grant funding, $1.5M industry revenue

Current focus: Independent methodology validation | Biomarker platform assessment | Trial design audits | Investor due diligence | Translational program development

Available for:

• 🔬 Biomarker validation audits (multi-site study review, confounding detection)
• 📋 Trial design & protocol review (endpoint validation, power analysis, regulatory risk)
• 🧪 Preclinical program guidance (IND-enabling strategy, translational planning)
• 💼 Investor due diligence (technical validation for biotech deals)
• 🏥 Digital health & diagnostic validation (algorithm validation, real-world performance)

Independent methodological investigation revealing severe confounding in high-profile proteomics study:

• 72% of published biomarkers failed geographic validation
• 99.9% accuracy predicting technical artifacts (tube type) vs. disease
• $15M+ estimated preventable follow-up costs

Key finding: Perfect confounding between diagnostic groups and technical factors created models learning artifacts instead of biology. Standard validation (98% pooled CV accuracy) missed this; geographic validation revealed deployment failure (77% AUC cross-site).

📄 Read full technical report with reproducible analysis →

• Platforms: Olink PEA, SomaScan, Metabolon, RNA-Seq (bulk/single-cell), WGS/WES
• Integration: WGCNA, MEGENA, MOFA, DIABLO, network-based approaches
• Validation: Geographic/leave-site-out CV, reverse prediction, stratified replication

• Diseases: ALS, Alzheimer's, Parkinson's/LBD, MS, metabolic syndrome
• Methods: Biomarker validation, patient stratification, clinical trial design, RWE studies
• Neurophysiology: EEG/ERP, fMRI, eye tracking, digital biomarkers

• Clinical: Trial design (Phase 1/2), endpoint selection, power analysis, CDISC standards
• Discovery: Target ID, MoA characterization, AI-enabled discovery
• Regulatory: IND-enabling studies, FDA submission strategy, biomarker qualification

• Languages: R (expert), Python, SQL, Stan
• ML/Stats: Statistical modeling, causal inference, longitudinal analysis, Bayesian methods
• Infrastructure: AWS/GCP, Docker, Shiny, reproducible pipelines (targets, renv)
• Specialties: Confounding detection, batch effect quantification, cross-site validation

All consulting projects deliver:

• Reproducible analysis code (R/Python, version-controlled)
• Transparent statistical frameworks
• Documented methodology with validation tests

65+ peer-reviewed publications | Full list on Google Scholar

Recent highlights:

• Kornilov, S. (2025). Confounding by geography and anticoagulant compromises proposed ALS diagnostic model and biomarkers: Reanalysis of Chia et al. (2025). Technical Report. Link

• Kornilov, S. (2025). When Algorithms Learn to Discriminate. Tech Policy Press, [Link](https://www.techpolicy.press/when-algorithms-learn-to-discriminate-the-hidden-crisis-of-emergent-ableism/

• Kornilov, S., Price, N., Gelinas, R., Brunkow, M, ... & Magis, A. (2024). Multi-Omic characterization of the effects of Ocrelizumab in patients with relapsing-remitting multiple sclerosis. Journal of the Neurological Sciences, 467, 123303. 10.1016/j.jns.2024.123303

• Heath, L., Earls, J., Magis, A., Kornilov, S., ... Price, N. (2022). Manifestations of Alzheimer's disease genetic risk in the blood are evident in a multiomic analysis in healthy adults aged 18 to 90. Scientific Reports, 12(6117). 10.1038/s41598-022-09825-2

• Kornilov, S., Lucas, ... & Magis, A. (2020). Plasma levels of soluble ACE2 are associated with sex, Metabolic Syndrome, and its biomarkers in a large cohort. Critical Care, 24, 452. 10.1186/s13054-020-03141-9

• Kornilov, S., Rakhlin, N., ... & Grigorenko, E.L. (2016). Genome-Wide Association and Exome Sequencing Study of Language Disorder in an Isolated Population. Pediatrics, 137(4). 10.1542/peds.2015-2469

View all publications →

Founder & Principal | Biostochastics LLC (2024-Present)
Independent biomarker validation and trial design consulting

Senior Research Scientist | Institute for Systems Biology (2019-2024)
Multi-omic studies, clinical collaborations (Bryleos, Genentech, Gilead, Thorne, Sanitarium)

Statistical Geneticist | Arivale Inc (2018-2019)
Precision wellness platform, longitudinal multi-omic analysis

Research Assistant Professor | University of Houston (2017-2018)

PhD, Experimental Psychology | University of Connecticut
Neurophysiological and genetic bases of developmental language disorder

PhD, Educational Psychology/Psychometrics | Moscow State University

Post-doctoral training in Molecular Genetics: Yale School of Medicine & Baylor College of Medicine (Duncan Scholarship in Molecular and Human Genetics)

Awards: Outstanding Doctoral Dissertation Award (SRCD), GoldenHelix Award for Best Research, Isabelle Liberman Award

For biotech companies:

• Biomarker validation before publication/trials
• Trial design audits (Phase 1/2 planning)
• Preclinical program guidance (IND-enabling strategy)

For investors:

• Technical due diligence on biomarker platforms
• Portfolio company methodology support
• Deal evaluation (multi-omics, diagnostics, digital health)

For healthtech/diagnostics:

• Algorithm validation and deployment feasibility
• Digital biomarker validation
• Regulatory pathway assessment

📧 Contact: sergey.kornilov@biostochastics.com
💼 LinkedIn: linkedin.com/in/sergey-kornilov

Available for consulting engagements and fractional advisory roles. Response time: <24 hours.