Payman Tohidifar

Hi, welcome to my page!

I'm a computational biologist and biomolecular engineer with over 14 years of academic and industry experience spanning molecular biology, genetics, computational science, and AI, all in pursuit of understanding and engineering biological systems.

During my PhD, I studied a fundamental question in biology: how do cells sense chemical signals from their environment and translate them into action? I used bacterial chemotaxis as a model system to study this, focusing on how bacterial chemoreceptors sense unconventional signals, from protons to small molecules like alcohols and aromatics to macromolecules like DNA, and encode them into a response. I combined computational systems biology and bioinformatics with experimental approaches, including advanced genetics, protein engineering, NGS, and structural biology, to answer these questions.

In my postdoctoral work, I moved from asking fundamental questions to application: engineering cells to transform low-value feedstocks into valuable small molecules through enzymatic pathways. This meant studying the physiology of these cells and designing tools to enable the enzymatic transformation of molecules into the final product, using omics analysis and advanced genetics. I carried this mission into industry, leading a cross-functional project to biosynthesize valuable small molecules, before shifting my focus to the synthesis and engineering of industrial enzymes. Watching AI and generative AI mature and prove themselves in biotech and pharma (such as AlphaFold, ESM, Evo, CodonFM, Boltz, and recently AI agents), I saw their potential to make biology more predictable, reduce silos across the R&D pipeline, and shorten discovery timelines. I drew on my computational background to start building toward that intersection myself.

That pursuit began 3+ years ago with the fundamentals of machine learning and has since grown into hands-on work with modern deep learning architectures, from CNNs and RNNs to transformers and biological foundation models, applied to protein function prediction, protein engineering, and gene regulation. It's been exciting to watch tech, biotech, and biopharma converge around these tools to tackle some of biology's hardest problems that impact human life and well-being. I'm now passionate about applying and building AI/ML tools that make biological discovery faster, more predictable, and more automated.

This site collects some of my projects with computational components, CV, and occasional writing on computational biology, synthetic biology, and AI.