Leveraging WES data to identify biomarker signatures

Overview

This case study highlights Excelra’s expertise in identifying biomarker signatures from WES data to support immunotherapy decision-making in oncology. A US-based pharmaceutical company partnered with Excelra to uncover biomarkers linked to efficacy and resistance in immuno-oncology using integrated whole exome sequencing, gene expression profiling, and advanced computational biology approaches supported by bioinformatics solutions.

Our client

The client is a US-based pharmaceutical company operating in the biopharma sector with a focus on immuno-oncology. Their research centered on Small Cell Carcinoma and Renal Cell Carcinoma, with a need to identify robust biomarker signatures predictive of immunotherapy response and resistance using large-scale genomic and transcriptomic datasets.

Client’s challenge

The client required a scalable, reproducible framework to identify clinically relevant biomarker signatures by integrating gene expression profiles with somatic mutation data. Traditional analytical workflows struggled to capture the complexity of tumor and tumor microenvironment interactions, limiting accurate interpretation of immunotherapy outcomes. A data-driven, systems-level approach was needed to extract biomarker signatures from WES data while ensuring reproducibility and alignment with evolving clinical evidence.

Client’s goals

The primary objective was to discover and validate biomarker signatures associated with immunotherapy efficacy and resistance. The client also sought optimized, validated analysis code and network models that could be reused internally, supporting long-term research programs and intellectual property development through scientific informatics services.

Our approach

Gene expression network construction

Excelra constructed gene co-expression networks using Weighted Gene Co-expression Network Analysis (WGCNA) to identify biologically meaningful modules from transcriptomic data. Multiple parameter combinations were tested to ensure robustness, and modules were correlated with clinical traits such as progression-free and overall survival. This approach aligns with best practices described in our AI-driven drug discovery analytics.

Module reduction and signature identification

Identified modules were reduced to biologically significant gene signatures using pathway knowledge, immune signatures, and elastic-net–based feature selection. Gene set enrichment analysis was performed using curated pathway collections, ensuring interpretability and relevance of the derived biomarker signatures from WES data.

Integration of somatic mutations and pathway context

Somatic mutation profiles from tumor-only WES data were integrated with baseline gene expression profiles. Advanced network-based algorithms such as DawnRank, DriverNet, and Paradigm-Shift were applied to understand functional impact of mutations. Pathway topology was augmented using curated resources and co-expression evidence, consistent with computational biology services.

Our solution

Consensus network modeling and validation

Excelra generated bootstrapped consensus networks to identify reproducible and robust modules. The analytical code was validated by reproducing published results from large immuno-oncology studies, ensuring scientific rigor and reproducibility in line with FAIR data principles.

Reusable analytics and knowledge assets

The final deliverables included annotated network modules, prioritized gene signatures, augmented pathway definitions, and filtered functional mutations. All analysis code was shared in notebook format, enabling continued internal use and integration with the client’s scientific data management systems.

Result

The project delivered validated biomarker signatures associated with immunotherapy efficacy and resistance across multiple cancer types. By combining WES data, transcriptomics, and network analytics, Excelra provided actionable insights into tumor biology and treatment response. The reusable analytical framework significantly accelerated downstream research while strengthening confidence in biomarker discovery outcomes.