Biotech researcher working where the life sciences meet data, AI and venture capital. MSc Bioscience Innovation and Enterprise at University College London (2026–27).

After three years at the bench at Skoltech, the Russian Academy of Sciences and Westlake University, I now study a different question: how capital decides which science becomes a product.

Do investors pick better biology? A reproducible open-data pipeline (Open Targets, ClinicalTrials.gov, NIH RePORTER) on 35,887 drug × indication programmes. It reproduces the genetic-support analysis of Minikel et al. (Nature, 2024) exactly and extends it to sponsor type: human genetic support predicts phase advancement about equally for small biotechs and top-20 pharma, yet small sponsors start fewer such programmes. Sponsors that raised private money (SEC Form D) do not pick better-supported targets either. Tested in CI, with every headline number checked against the model output.

• Separating venture-backed from other privately financed biotechs: SEC Form D private placements are now linked to trial sponsors; next come a US-only comparison and the venture investors named on each filing
• An AI tool that would turn static PDF lab protocols into interactive checklists with timers, a protocol Q&A assistant and auto-generated reports, growing out of the Python pipelines I wrote at Skoltech
• BioSauna, a bioactive steam-therapy startup that won ~$16k in non-dilutive university grants

• Skoltech: bacterial defence against phages. First experimental evidence of restriction alleviation in a type II restriction–modification system; Python pipelines that process metagenomic data and generate monthly research reports
• Russian Academy of Sciences: raised the yield of the anti-TNF mini-antibody MYSTI-2 from ~120 to ~200 µg/mL at 24 h by switching to auto-induction medium; presented at Lomonosov-2025 (MSU)
• Westlake University: full-ride Frontiers of Life Sciences program (~3% acceptance); protein design and purification
• ZOZH (Zozh Foods), New York, remote: market and product strategy for a personalised-nutrition startup; analysed U.S. supplement demand and the effect of GLP-1 drugs on eating behaviour
• R&D Pharma, Monaco, remote: literature reviews and experimental-design input for a medical device for skin lesions
• MSc Bioinformatics and Molecular Biotechnology, HSE University (2025–26); BSc Biotechnology, Sechenov University (GPA 4.81/5.0; thesis defended with highest distinction)

Data & AI: Python (pandas, statsmodels, RapidFuzz, matplotlib), OpenAI API, n8n, pytest, GitHub Actions Bioinformatics: sequence analysis, plasmid design (SnapGene, Benchling), protein structure (PyMOL), basic ML for sequence and protein analysis, statistics for expression data Wet lab: CRISPR-Cas9, λ-Red recombineering, generalized transduction, cloning, recombinant protein expression and purification (IMAC, ion exchange, gel filtration), SDS-PAGE, Western blot, ELISA, phage assays

• Email: [email protected]
• LinkedIn: linkedin.com/in/mark-lapunov