Teaching a course entitled “Protein Principles in Biotech”, one of the core courses in the Master’s level Biotechnology Program at Northeastern. A highly practice-oriented program, designed to prepare graduates for success in biotech.

Consulting in Chemistry and Machine Learning. I'm the only founder of this LLC company.

Optimization of chemical compounds and chemical reactions with ML: • I designed, trained and deployed new ML models of various architectures (including, but not limited to: 3D CNN, LSTM/GRU/multi-head-attention-based, FCNN, SVM, RF), and engineered appropriate ML descriptors (quantum-chemistry-, 3D-geometry-, structure-based, etc.). • Carried out projects on various chemical and physico-chemical property predictions: potency (protein-ligand interactions), log D (lipophilicity), pKa (acidity), regioselectivity, reaction yield, etc. • Closely collaborated with project teams to iteratively optimize target properties of drug discovery leads and their synthetic schemes in a computationally guided way.Generative Chemistry: • Developed a new ML model and internalized several models from the literature to efficiently generate high-quality virtual structures. • Worked with project teams to create tailor-made computation workflows. • Hundreds of compounds I suggested were synthesized on four drug design projects, some proved to be highly active.Quantum Chemistry modeling: • Built, optimized and deployed a regioselectivity prediction model, • Explored opportunities for reactivity prediction (in the context of late stage functionalization) based on transition state modeling.Computational infrastructure: • Extensively used Python (PyTorch, scikit-learn, etc.), Bash, SLURM, Git in my work. • Created and maintained Jupyter notebooks so that colleagues without computational background could use our models themselves.Mentored 3 interns: • An undergrad student from U Toronto: a project on potency prediction with ML. • An undergrad student from MIT: generative chemistry. • A grad student from MIT: quantum chemical modeling.

At Stanford, I developed a deep learning framework for accurate computation of electron densities and energies of organic molecules. Computations with this method are more accurate and take much less time than corresponding DFT calculations. I also initiated and worked on a project on the structure, dynamics and function of NMDA receptors, which exhibit complex allostery and play important roles in learning, memory formation, as well as various neurological disorders. It is difficult to study the structure and function of NMDARs at atomic resolution due to their large size and fast dynamics. Computer simulations offer unique opportunities for studying these receptors.Publications at Stanford:• Sinitskiy, A. V. & Pande, V. S. (2019) Physical machine learning outperforms 'human learning' in Quantum Chemistry. arXiv:1908.00971.• Sinitskiy, A. V. & Pande, V. S. (2018) Deep neural network computes electron densities and energies of a large set of organic molecules faster than Density Functional Theory (DFT). arXiv:1809.02723.• Sinitskiy, A. V. & Pande, V. S. (2018) Computer simulations predict high structural heterogeneity of functional state of NMDA receptors. Submitted. Biophys. J., 115, 841-852.• Sinitskiy, A. V. & Pande, V. S. (2017) Theoretical restrictions on longest implicit timescales in Markov state models of biomolecular dynamics. J. Chem. Phys., 148, 044111.• Sinitskiy, A. V. & Pande, V. S. (2017) Simulated dynamics of glycans on ligand-binding domain of NMDA receptors reveals strong dynamic coupling between glycans and protein core. J. Chem. Theory Comput., 13, 5496-5505.• Sinitskiy, A. V., Stanley, N. H., Hackos, D. H., Hanson, J. E., Sellers, B. D. & Pande, V. S. (2017) Computationally discovered potentiating role of glycans on NMDA receptors. Sci. Rep., 7, 44578.

At UChicago, I developed new computational tools and methods and applied them to various molecular and biomolecular systems on scales inaccessible to experimental or previously existing computational techniques. My Ph.D. thesis was entitled “Multiscale modeling of large biomolecular systems,” and contained applications to actin filaments and other systems of biological importance.• Sinitskiy, A. V. & Voth, G. A. (2017) Quantum mechanics / coarse-grained molecular mechanics (QM/CG-MM). J. Chem. Phys., 148, 014102.• Madsen, J.*, Sinitskiy, A. V.*, Li, J.* & Voth, G. A. (2017) Highly coarse-grained representation of transmembrane proteins. J. Chem. Theory Comput., 13, 935-944. [* Authors contributed equally]• Hocky, G. M., Baker, J. L., Bradley, M. J., Sinitskiy, A. V., De La Cruz, E. M. & Voth, G. A. (2016) Cations stiffen actin filaments by adhering a key structural element to adjacent subunits. J. Phys. Chem. B, 120, 4558-4567.• Sinitskiy, A. V. & Voth G. A. (2015) A reductionist perspective on quantum statistical mechanics: Coarse-graining of path integrals. J. Chem. Phys., 143, 094104. • Davtyan, A., Dama, J. F., Sinitskiy, A. V. & Voth G. A. (2014) The theory of ultra-coarse-graining. 2. Numerical implementation. J. Chem. Theory Comput., 10, 5265-5275.• Jang, S., Sinitskiy, A. V. & Voth, G. A. (2014) Can the ring polymer molecular dynamics method be interpreted as real time quantum dynamics? J. Chem. Phys., 140, 154103.• Dama, J. F.*, Sinitskiy, A. V.*, McCullagh, M., Weare, J., Roux, B., Dinner, A. R. & Voth, G. A. (2013). The theory of ultra-coarse-graining. 1. General principles. J. Chem. Theory Comput., 9, 2466-2480.• Sinitskiy, A. V. & Voth, G. A. (2013). Coarse-graining of proteins based on elastic network models. Chem. Phys., 422, 165-174.• Sinitskiy, A. V., Saunders, M. G. & Voth, G. A. (2012). Optimal number of coarse-grained sites in different components of large biomolecular complexes. J. Phys. Chem. B, 116, 8363–8374.