Research Areas

The Jain Laboratory was part of the University of California, San Francisco for 22 years (1999-2021). It is now Jainlab LLC, established to maintain and expand the academically focused distribution of data relating to the following research areas:

  • Small-molecule conformer generation, including complex macrocycles
  • Molecular docking and protein binding site comparison and analysis
  • Three-dimensional molecular similarity and multiple ligand alignment
  • Prediction of affinity and pose for biologically interesting molecules
  • Improvements to modeling X-ray density, particularly by employing conformational ensembles 

The work represents a unique approach in CADD by respecting physics, chemistry, and biology while also taking machine learning and parameter optimization seriously.

Tools Module: ForceGen methodology

Docking Module: Virtual screening and highly accurate pose prediction

Similarity Module: eSim method for virtual screening and multiple ligand alignment

Affinity Module: Machine learning using QuanSA

What’s New?

JMC: Bound Ligand Strain
January 27, 2023

We have published a comprehensive study of bound ligand strain in J. Med. Chem., with our colleagues from Merck. Ligand strain, when properly estimated using real-space refinement, is generally quite low. But it grows more rapidly than ligand size and follows a simple distributional model. The super-linearity of ligand strain w/r/t size explains a part of the challenge in designing larger ligands that are highly efficient.

JCAMD: The eSim  Similarity Method
October 24, 2019

We have published, with Stephen Johnson, a new 3D molecular similarity method (called eSim) that directly incorporates Coulombic field comparison with surface-shape and hydrogen-bonding comparison. It is both faster and more accurate that commonly used alternatives, both for virtual screening and pose prediction. The paper is entitled “Electrostatic-field and surface-shape similarity for virtual screening and pose prediction.”

JMC: xGen Papers
March 17, 2021

We have published a second paper with colleagues from Merck looking at peptide macrocycle strain energetics in the context of xGen refined conformational ensembles. This follows the paper entitled “XGen: Real-Space Fitting of Complex Ligand Conformational Ensembles to X‐ray Electron Density Maps.” We show that conformational ensembles, without atom-specific B-factors, are better models for ligands in terms of both fit to X-ray density and strain energy.

JCAMD: QuanSA Affinity Prediction Method
December 10, 2021

A focused machine-learning approach to inducing binding site models with QuanSA is shown to be complementary and synergistic with predictions from FEP+ across 17 targets. The paper reports significant reductions in prediction errors and consistent improvements in compound rank-ordering by combining the results of QuanSA and FEP+. The introductory paper offers additional details about this novel multiple-instance machine-learning method for binding affinity and pose prediction.