Wei Zhang

Cross-functional leader of solid tumor precision discovery projects; Impact target discovery strategies in oncology

• Led cross-functional teams for precision target discovery in oncology. • Drove computational and experimental strategies in disease understanding, patient stratification, and tumor intrinsic/extrinsic target identification. • Oversaw the development of organoid models and perturbation assays for target validation. • Developed computational strategies for target identification (e.g. oncogenic drivers, synthetic lethality) • Mapped pre-clinical models to patient states for large-scale genetic/compound screenings. • Managed and mentored a team of computational biologists (scientist to principal scientist level)

• Developed a causal inference method (by extending MR-Egger) for identifying disease-specific drivers linked to mutations and patient outcomes. This method became a foundation for new target discovery strategies in the organization. (BioRxiv 2025) • Developed an analytical framework (using GLM and RF) to link compound features to their proteomics activities, enabling hit nomination for molecular glue compounds by the early discovery organization.

• Led computational research during early development of two PROTAC programs at BMS in hematology and solid tumors (AR-LDD and BCL6-LDD), currently in clinical development. • Developed AI/ML frameworks (using VAE, PLS, GLM) for integrating multi-omics data types (cell painting, transcriptomics, proteomics) to generate mechanistic insights of molecular glue compounds.

• Integrated whole genome sequences and transcriptomic data to identify kidney disease-specific expression quantitative trait loci (eQTLs). • Inferred kidney disease severity from patients' longitudinal clinical data using a Bayesian hierarchical linear model. Conducted association tests between genetic variants and disease severity. • Transcriptome-wide association study (TWAS): Trained elastic-net models to impute gene expression from genetic variants; conducted association tests between imputed expression and disease phenotype. • Single cell RNA-Seq analyses for kidney organoids. • Presented posters on Goldfinch’s Kidney Genome Atlas at ASHG 2018 and ASN 2018.

• Developed a somatic eQTL analysis to interpret the noncoding mutations in cancer. (Nature Genetics, 2018) • By implementing a random walk model, we identified multiple oncogenic pathways promoted by HPV interactions that phenocopy recurrent mutations in cancer. (Cancer Discovery, 2018) • Developed an algorithm by adapting the Supervised Random Walk with a novel loss function designed specifically for cancer classification. (ISMB and Bioinformatics, 2018) GitHub: https://github.com/wzhang1984/NBSS

• Developed a framework to analyze 2 million variants of synthetic enhancers in living embryos, and found that sequences flanking transcription factor binding sites are important for gene expression. (Science 2015; PNAS 2016) • Examined the ChIP-seq data of RNA Pol II and ELAV, and provided the first evidence for a regulatory link between promoter-proximal Pol II pausing and alternative polyadenylation. (Molecular cell 2015)