Shan Chen

. Established GPU cluster infrastructure to support enterprise-scale AIDD computational workflows. . Initiated construction of a scientific data lakehouse to standardize affinity, druggability, and PK data for future AI model training and reuse. . Launched peptide de novo design and reference-based redesign workflows for agonistic and antagonistic peptides by combining generative AI with molecular dynamics and other physics-based guardrails. . Delivered antagonistic peptide designs advancing through experimental validation. . Developed an integrated AI + physics pipeline for small-molecule pose prediction, enabling accurate identification of core structures and activity cliffs validated through wet-lab design. . In-licensed and deployed an in-house FEP platform for affinity prediction across small molecules and cyclic peptides.

. Led AI-powered peptide design platform development using reinforcement learning, diffusion, and transformer-based workflows. . Built structure-guided design workflows integrating protein structure prediction with multi-objective optimization across affinity, specificity, and druggability. . Directed development of LLM-agent capabilities for target validation and scientific decision support. . Supported external partnering and licensing through bioinformatics diligence, indication expansion analyses, and biomarker strategy. . Expanded AI applications in imaging and translational discovery workflows.

Building state-of-art bioinformatics suites for immune-oncology drug target identificaiton. Coroborating with robotic lab to advance high throughput drug screening system.

Lead cross-fuction groups to build state-of-art automatic pipeline for rare variant interpretation and rare disease diagnosis.

1. Lead projects to identify genomic variations associated with drug response, adverse events using state-of-art Genome-Wide Association Study (GWAS). 2. Stratify patients in terms of disease risk, drug response or toxicity using Polygenic Risk Score, based on Bayesian or Machine Learning algorithms to summarize genome-wide variant contribution to various traits.

• Leading bioinformatician for developing causal reasoning pipeline for drug target identification utilizing patient transcriptome data • Lead work in drug-target benchmark curation and evaluation • Delivered hypothesis generation for internal disease programs with patient omics data • Prototyped single-cell RNA-seq, co-expression analyses for hypothesis generation • Managed and interacted transcriptome data deployed to Google Cloud and BigQuery

• Lead bioinformatician for Undiagnosed Disease Network at Baylor College of Medicine, responsible for pipeline development, variant prioritization, clinical case analyses, data aggregation and query. • Developed RNA-seq pipeline which successfully automated identification of abnormal gene expression, aberrant isoforms, allelic expression and uniquely regulated pathways. • Developed WGS pipeline focusing on non-coding variant annotation and integrated with RNA-seq data for variant prioritization. • Deployed RNA-seq pipeline to cloud-computing platform i.e. SevenBridges for scalability, reproducibility and collaboration. • Built RNA-seq reference database with Google BigQuery for SQL management of sequencing data, query and statistical analyses. • Provide statistical consultation to collaborators for study design, power analysis and model building for mixed or longitudinal data.

• GWAS and burden analysis to identify susceptibility genes for Pheochromocytoma and Paraganglioma (PCC/PGL) with whole exome sequencing (WES) using VAAST. • Integrative genomics analysis to validate PCC/PGL genes by combining somatic variants, CNAs and transcriptome using TCGA data. • GWAS study to identify single nucleotide polymorphisms for Crohn’s disease with targeted sequencing.

• ChIP-seq analysis incorporating pathway and motif analyses to identify new transcription regulatory sequences for Notch signaling • Identify somatic mutations associated with osteosarcoma with tumor-normal paired WES • Designed, generated and characterized genetic murine models to study Notch signaling in cartilage. • Designed and implemented ChIP-seq, RNA-seq and microarray experiments to investigate novel transcription regulation mechanism of Notch. • Performed cellular and molecular study to validate candidate genes identified through ChIP-seq.