Siddhant Rath sidxz

About Me

I'm a software and machine learning engineer at Texas A&M AgriLife Research, working alongside structural biologists, medicinal chemists, and computational scientists on early-stage drug discovery.

📍 Austin, TX · M.S. CS, Texas A&M

Selected Publications

Year	Publication	Journal
2026	DocuStore: Domain Constrained Retrieval Augmented Generation for Drug Discovery Documents	Preprint
2026	CAGE Fusion: Deep Learning for Nuisance Compound Detection Using a Gated Co-Attention Graph Embedding Model	Journal of Cheminformatics, Springer Nature
2025	Integrating Chemical, Genetic, and Feasibility Assessments for Anti-Tubercular Target Validation	EMBO Molecular Medicine, Springer Nature
2023	DAIKON: A Data Acquisition, Integration, and Knowledge Capture Web App for Target-Based Drug Discovery	ACS Pharmacology & Translational Science, American Chemical Society

Production Applications

Applications I lead or have led from architecture and development through production deployment, supporting drug discovery teams across academic, nonprofit, and pharmaceutical organizations.

	DocuStore · docustore.io · Publication · GitHub A chemistry-aware, domain-constrained RAG and document intelligence platform for drug discovery. DocuStore extracts and connects chemical structures, compounds, biological entities, bioactivity measurements, and scientific text to enable evidence-grounded search and question answering across complex research documents.
	CAGE-Fusion · Publication · GitHub · Featured In ( Drug Target Review · Phys.org · The Microbiologist · News-Medical ) A deep learning model for molecular property prediction, specialized in identifying nuisance compounds that can confound early-stage drug discovery assays. CAGE-Fusion predicts aggregation, luciferase inhibition, chemical reactivity, and promiscuity using learned molecular representations.
	DAIKON AI & Curator Studio · Closed user group testing · Coming soon A collaborative drug discovery platform for scientific data curation, AI/ML prediction, analysis, and project management across the early discovery lifecycle, from genes and targets through screening, hit assessment, portfolio management, and post-portfolio studies. DAIKON AI & Curator Studio is the next generation of DAIKON, extending a platform used by the TB Drug Accelerator consortium since 2022 to support collaborative tuberculosis drug discovery.
	ChemCellar · chemcellar.com · GitHub An open-source compound registration, management, and screening platform for research organizations. ChemCellar provides a centralized environment for managing molecules, assays, protocols, compound collections, screening data, and research projects.
	DAIKON · saclab.github.io/daikon · GitHub · Publication An open-source framework for collaborative, target-based drug discovery. DAIKON integrates targets, screens, hits, compounds, and project portfolios into a unified system that captures the progression of discovery programs and enables scientific teams to curate, analyze, visualize, and share project data. DAIKON has been used in production within the TB Drug Accelerator (TBDA), a Gates Foundation initiative, since 2022, supporting collaborative drug discovery across academic and pharmaceutical research organizations.

Other Projects

Duar _{Self-hosted authorization for multi-tenant apps: workspaces, RBAC, and per-resource permissions on top of any OIDC provider.} _{duar.io · GitHub · Docs}	structflo.cser _{Chemical structure and label extraction from scientific documents.} _{GitHub · PyPI}	structflo.ner _{Zero-config named entity recognition for drug discovery, chemistry, and biological sciences.} _{GitHub · PyPI}
_{Explore public evidence on M. tuberculosis H37Rv drug targets: genes, structures, compound activity, screens, resistance, drugs and trials.} _GitHub	Molseal _{Encrypted molecular storage with RDKit: AES-256-GCM records in SQL, with substructure, similarity and exact-key search.} _{GitHub · PyPI}