Biology subjects

Davtyan, A.

Publications and source records attributed to Davtyan, A..

5 recordsLinked to original sources

bioRxiv2026

Overcoming the accuracy-generalization tradeoff in docking and scoring for prospective virtual screening

Virtual screening promises access to tens of billions of synthetically accessible, diverse compounds, yet it is rarely used as a primary hit-discovery strategy in contemporary drug-discovery campaigns. We argue that this gap reflects the real-world underperformance of the underlying docking and scoring methods: classical docking is generalizable but limited in accuracy by simple functional forms and parsimonious parameterization, whereas recent machine-learning approaches are highly expressive but do not generalize well to novel molecules and pockets, their reported accuracy often inflated by train-test leakage. To address these challenges, we introduce DODock and DOScore, docking and scoring ML/physics hybrid frameworks that are also highly expressive, yet generalize much better out of distribution compared with the prior ML approaches. This generalization has been prospectively tested in several ways. First, DODocks blind prediction of a drug candidate binding to PCSK9 was compared to the crystal structure that was subsequently solved, recovering the binding pose to 1.2 [A] RMSD. We also used DODock and DOScore in prospective virtual screening campaigns against four therapeutic targets, spanning an ectoenzyme (CD73), a kinase (IRAK4), an extended-substrate protease (FXI), and an allosteric inhibition of protein-protein interface (IL17). These screens yielded many chemically novel, biochemically and cellularly active inhibitors. In the case of CD73, which is a historically challenging target for virtual screening, our screen resulted in a roughly hundredfold improvement in hit rate over a recent machine-learning screen. Our results indicate that the apparent ceiling in the accuracy of virtual screening that seemed to have somewhat plateaued in the last two decades is not fundamental, and that structure-based interrogation of ultralarge chemical space may eventually become a credible primary route to novel chemical matter.

biophysics↗

bioRxiv2025

AmyloDeep: pLM-based ensemble model for predicting amyloid propensity from the amino acid sequence

Amyloids are predominantly {beta}-sheet-rich, stable protein structures that can maintain their presence in the human body for multiple years. Amyloid protein aggregates contribute to the development of multiple neurodegenerative diseases, such as Alzheimers, Parkinsons, and Huntingtons, and are involved in different vital functions, such as memory formation and immune system function. Here, we used advanced machine learning and deep learning techniques to predict amyloid propensity from the amino acid sequence. First, we aggregated labeled amino acid sequence data from multiple sources, obtaining a roughly balanced dataset of 2366 sequences for binary classification. We leveraged that data to both fine-tune the ESM2 model and to train new models based on protein embeddings from ESM2 and UniRep. The predictions from these models were then unified into a single soft voting ensemble model, yielding highly robust and accurate results. We further made a tool where users can provide the amino acid sequence and get the amyloid formation probabilities of different segments of the input sequence. Users can access the light version of AmyloDeep through the web server at https://amylodeep.com/, and the full model is available as a Python package at https://pypi.org/project/amylodeep/. Graphical abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=47 SRC="FIGDIR/small/676495v1_ufig1.gif" ALT="Figure 1"> View larger version (13K): org.highwire.dtl.DTLVardef@64615aorg.highwire.dtl.DTLVardef@339a7corg.highwire.dtl.DTLVardef@1e36e14org.highwire.dtl.DTLVardef@4fe717_HPS_FORMAT_FIGEXP M_FIG C_FIG

bioinformatics↗

bioRxiv2023

Discovery of a cryptic pocket in the AI-predicted structure of PPM1D phosphatase explains the binding site and potency of its allosteric inhibitors

Virtual screening is a widely used tool for drug discovery, but its predictive power can vary dramatically depending on how much structural data is available. In the best case, crystal structures of a ligand-bound protein can help find more potent ligands. However, virtual screens tend to be less predictive when only ligand-free crystal structures are available, and even less predictive if a homology model or other predicted structure must be used. Here, we explore the possibility that this situation can be improved by better accounting for protein dynamics, as simulations started from a single structure have a reasonable chance of sampling nearby structures that are more compatible with ligand binding. As a specific example, we consider the cancer drug target PPM1D/Wip1 phosphatase, a protein that lacks crystal structures. High-throughput screens have led to the discovery of several allosteric inhibitors of PPM1D, but their binding mode remains unknown. To enable further drug discovery efforts, we assessed the predictive power of an AlphaFold-predicted structure of PPM1D and a Markov state model (MSM) built from molecular dynamics simulations initiated from that structure. Our simulations reveal a cryptic pocket at the interface between two important structural elements, the flap and hinge regions. Using deep learning to predict the pose quality of each docked compound for the active site and cryptic pocket suggests that the inhibitors strongly prefer binding to the cryptic pocket, consistent with their allosteric effect. The predicted affinities for the dynamically uncovered cryptic pocket also recapitulate the relative potencies of the compounds ({tau}b=0.70) better than the predicted affinities for the static AlphaFold-predicted structure ({tau}b=0.42). Taken together, these results suggest that targeting the cryptic pocket is a good strategy for drugging PPM1D and, more generally, that conformations selected from simulation can improve virtual screening when limited structural data is available.

biophysics↗

bioRxiv2020

Mesoscopic Liquid Clusters Represent a Distinct Condensate of Mutant p53

The oncogenic properties of mutant p53 have been ascribed to destabilization of the p53 conformation, followed by aggregation into insoluble fibrils. Here we combine immunofluorescent 3D confocal microscopy of breast cancer cells expressing the p53 mutant Arg248Gln (R248Q) with light scattering from solutions of the purified protein and molecular simulations to probe the mechanisms that govern phase behaviors of the mutant across multiple length scales, from cellular to molecular. We establish that p53 R248Q forms mesoscopic protein-rich clusters, an anomalous liquid phase with several unique properties. We demonstrate that the clusters host and facilitate the nucleation of amyloid fibrils. The distinct characteristics of the clusters of R248Q and wild-type p53 and theoretical models indicate that p53 condensation into clusters is driven by the structural destabilization of the core domain and not by interactions of its extensive disordered region. Two-step nucleation of mutant p53 amyloids suggests means to control fibrillization and the associated pathologies through modifying the cluster behaviors. In a broader context, our findings exemplify interactions between distinct protein phases that activate complex physicochemical mechanisms operating in biological systems.

biophysics↗

bioRxiv2018

Insights into the cooperative nature of ATP hydrolysis in actin filaments

Actin filaments continually assemble and disassemble within a cell. Assembled filaments \"age\" as a bound nucleotide ATP within each actin subunit quickly hydrolyzes, followed by a slower release of the phosphate Pi, leaving behind a bound ADP. This subtle change in nucleotide state of actin subunits affects filament rigidity as well as its interactions with binding partners. We present here a systematic multiscale ultra-coarse-graining (UCG) approach that provides a computationally efficient way to simulate a long actin filament undergoing ATP hydrolysis and phosphate release reactions, while systematically taking into account available atomistic details. The slower conformational changes and their dependence on the chemical reactions are simulated with the UCG model by assigning internal states to the coarse-grained sites. Each state is represented by a unique potential surface of a local heterogeneous elastic network. Internal states undergo stochastic transitions that are coupled to conformations of the underlying molecular system. The UCG model reproduces mechanical properties of the filament and allows us to study whether fluctuations in actin subunits produce cooperative aging in the filament. Our model predicts that nucleotide state of neighboring subunit significantly modulates the reaction kinetics, implying cooperativity in ATP hydrolysis and Pi release. We further systematically coarse-grain the system into a Markov state model that incorporates assembly and disassembly, facilitating a direct comparison with previously published models. We find that cooperativity in ATP hydrolysis and Pi release significantly affects the filament growth dynamics only near the critical G-actin monomer concentration, while both cooperative and random mechanisms show similar growth dynamics far from the critical concentration. In contrast, filament composition in terms of the bound nucleotide distribution varies significantly at all monomer concentrations studied. These results provide new insights into the cooperative nature of ATP hydrolysis and Pi release and the implications it has for actin filament properties, providing novel predictions for future experimental studies.

biophysics↗