Atluri, G. | Search bioRxiv

bioRxiv2025

Quantum and Classical Graph Convolutional Neural Networks for Protein Ligand Dissociation Constant Prediction

How long a drug stays bound to its target - the residence time - is now recognized as a stronger in vivo efficacy driver than binding affinity alone. Yet current machine learning (ML) models for dissociation kinetics (koff) ignore two critical sources of structure information: (1) the change in protein-ligand geometry across time, and (2) the ability to represent complex spatial interactions with fewer parameters. We extend a state-of -the -art spatial GNN for protein-ligand complexes with two innovations. First, we introduce a 2-timestep GCN+GRU model that learns structure changes before and after short molecular dynamics simulations. Second, we compress the model head using variational quantum circuits, preserving expressivity while removing 66% of parameters. On the full PDBbind-koff-2020 benchmark, temporal integration improves predictive accuracy, and quantum compression matches the full classical models accuracy at a fraction of its size. Our results show that time and quantum structure are underexplored, high-leverage axes for advancing kinetic ML models used in drug design.

bioinformatics↗

bioRxiv2019

Resolving single-cell heterogeneity from hundreds of thousands of cells through sequential hybrid clustering and NMF

The rapid proliferation of single-cell RNA-Sequencing (scRNA-Seq) technologies has spurred the development of diverse computational approaches to detect transcriptionally coherent populations. While the complexity of the algorithms for detecting heterogeneity have increased, most existing algorithms require significant user-tuning, are heavily reliant on dimensionality reduction techniques and are not scalable to ultra-large datasets. We previously described a multi-step algorithm, Iterative Clustering and Guide-gene selection (ICGS), which applies intra-gene correlation and hybrid clustering to uniquely resolve novel transcriptionally coherent cell populations from an intuitive graphical user interface. Here, we describe a new iteration of ICGS that outperforms state-of-the-art scRNA-Seq detection workflows when applied to well-established benchmarks. This approach combines multiple complementary subtype detection methods (HOPACH, sparse-NMF, cluster \"fitness\", SVM) to resolve rare and common cell-states, while minimizing differences due to donor or batch effects. Using data from the Human Cell Atlas, we show that the PageRank algorithm effectively down samples ultra-large scRNA-Seq datasets, without losing extremely rare or transcriptionally similar distinct cell-types and while recovering novel transcriptionally unique cell populations. We believe this new approach holds tremendous promise in reproducibly resolving hidden cell populations in complex datasets.\n\nHighlightsO_LIICGS2 outperforms alternative approaches in small and ultra-large benchmark datasets\nC_LIO_LIIntegrates multiple solutions for cell-type detection with supervised refinement\nC_LIO_LIScales effectively to resolve rare cell-states from ultra-large datasets using PageRank sampling with a low memory footprint\nC_LIO_LIIntegrated into AltAnalyze to enable sophisticated and automated downstream analysis\nC_LI

bioinformatics↗