Leo Gendelev

Grad Student; Fletcher Jones Fellow

Leo came into the micro-world of molecules and cells at UCSF from the macro-world of computational astrophysics at UC Santa Cruz where he obtained his B.Sc. degree running magneto-hydrodynamic simulations of star formation and analyzing pulsar gamma ray emissions. He began his adventures at UCSF with an interest in protein design and structural modeling, but since then he got even more excited about research at the intersection of deep-learning and drug discovery in the context of high-throughput phenotypic screening. Several of his projects have involved computational target deconvolution and polypharmacology in the context of specific neuroactive phenotypes in zebrafish (eg. antipsychotics, sedatives, etc…) using statistics and cheminformatics based approaches such as the Similarity Ensemble Approach (SEA). Over the last several years he has helped initiate and drive the DeepFish collaboration together with the Kokel Lab - using various deep-learning techniques such as variational autoencoders and siamese neural networks to improve our ability to extract relevant features from zebrafish behavioral videos - and build better distance metrics for drug-similarity and drug discovery. He strongly believes in deriving insights from structural biophysics and molecular biology at every step along the way, and most importantly, unraveling the “black-box” of deep learning - using machine learning to actually test specific biological hypothesis and derive actionable explanations.

Lab papers

  1. Zebrafish behavioural profiling identifies GABA and serotonin receptor ligands related to sedation and paradoxical excitation.

    McCarroll MN, Gendelev L, Kinser R, Taylor J, Bruni G, Myers-Turnbull D, Helsell C, Carbajal A, Rinaldi C, Kang HJ, Gong JH, Sello JK, Tomita S, Peterson RT, Keiser MJ, Kokel D. Nat Commun. 2019 Sep 9.

  2. A Simple Representation of Three-Dimensional Molecular Structure.

    Axen SD, Huang XP, Caceres EL, Gendelev L, Roth BL, Keiser MJ. J Med Chem. 2017 Sep 14.

  3. Zebrafish behavioral profiling identifies multitarget antipsychotic-like compounds.

    Bruni G, Rennekamp AJ, Velenich A, McCarroll M, Gendelev L, Fertsch E, Taylor J, Lakhani P, Lensen D, Evron T, Lorello PJ, Huang XP, Kolczewski S, Carey G, Caldarone BJ, Prinssen E, Roth BL, Keiser MJ, Peterson RT, Kokel D. Nat Chem Biol. 2016 Jul.

  4. Leveraging Large-scale Behavioral Profiling in Zebrafish to Explore Neuroactive Polypharmacology.

    McCarroll MN, Gendelev L, Keiser MJ, Kokel D. ACS Chem Biol. 2016 Apr 15.