Abhinaw Kumar
Abhinaw Kumar
Abhinaw Kumar
Postdoctoral Research Associate, University of Arkansas • PhD, Computational Chemistry (University of Utah) • Machine learning + molecular simulation
I am a computational chemist specializing in molecular simulation, statistical mechanics, self-assembly, nucleation, protein biophysics, and machine-learning-based multiscale modeling. Although the systems I study have evolved — from nanoparticles and zeolites to intrinsically disordered proteins and biomolecular condensates — my approach has stayed consistent: build simplified molecular representations that retain the essential physics, use them to reach processes that are hard to access at full atomistic resolution, and develop methods that connect reduced descriptions back to detailed ones. My current work uses probabilistic normalizing-flow models to reconstruct atomistic ensembles from coarse-grained configurations of diphenylalanine, combining generative modeling with molecular mechanics and statistical-mechanical reweighting so that generated structures can be evaluated against thermodynamic distributions.
Generative modeling Multiscale simulation Statistical mechanics Nucleation & self-assembly IDPs & condensates Porous materials

Research Interests

  • AI and data science for chemistry: probabilistic generative modeling, coarse-grained-to-atomistic mapping, and multiscale simulation
  • Molecular simulation and statistical mechanics of self-assembly, nucleation, polymorphism, and phase transitions
  • Intrinsically disordered proteins, protein aggregation, biomolecular condensates, and protein fibrils
  • Porous materials and soft matter: zeolites, mesophases, nanoparticle assembly, and inverse-design modeling

Research Trajectory

2024 – present · Postdoctoral Research Associate — University of Arkansas (Monroe Group)
Machine-learning-based probabilistic backmapping for multiscale modeling of diphenylalanine (FF) peptide assembly: designing, training, and validating autoregressive normalizing flows with structural, energetic, and statistical metrics, molecular-dynamics refinement, and ensemble reweighting.
2023 – 2024 · Postdoctoral Research Associate — Texas A&M University (Mittal Group)
Coarse-grained simulations of protein–DNA (HP1–DNA) condensates to uncover mechanisms of chromatin compaction, using HOOMD-blue and HPS models to analyze thermodynamic stability and diffusion dynamics.
2019 – 2023 · Postdoctoral Research Associate — UT Austin (Thirumalai Group)
Intrinsically disordered proteins and liquid–liquid phase separation; force-field optimization validated against experimental SAXS profiles; phase behavior, fibril formation, and phosphorylation effects in FUS and FUS-LC, including distinct fibril-core morphologies and sequence-dependent stability.
2012 – 2018 · PhD, Computational Chemistry — University of Utah (Molinero Group)
Thermodynamics of nanoparticle self-assembly and phase transitions (LAMMPS); nucleation of ordered phases, transition path sampling, and simulation of zeolite formation. Introduced the Kumar–Molinero mesophase model and the mesophase-mediated picture of zeolite nucleation.
2007 – 2012 · BS–MS Dual Degree in Chemistry, with Distinction — IISER Pune
Dissertation: Interaction of Polyethyleneimine with Membrane Bilayer at Different pH: a Molecular Dynamics Study — atomistic simulations of pore formation in polymer–lipid bilayer systems (GROMACS).

Education

University of Utah — PhD in Computational Chemistry
Dissertation: Self-Assembly of Mesophases and Their Role in the Nucleation and Polymorph Selection of Zeolites. Advisor: Prof. Valeria Molinero.
IISER Pune — BS–MS Dual Degree in Chemistry (Distinction)
INSPIRE Fellow (2008–2012); Academic Excellence Award (2010). Undergraduate research internships in India, Germany, and the UK.

Peer Review & Scientific Service

Peer reviewer
19 review assignments covering 17 manuscripts for international journals including Molecular Simulation, Journal of Molecular Liquids, Frontiers in Soft Matter, Polymers, International Journal of Molecular Sciences, Molecules, Biophysica, Crystals, and Pharmaceutics.
Judge, Texas Junior Science and Humanities Symposium
Evaluated original student research projects in chemistry and related physical sciences, selecting awardees on scientific rigor, originality, and technical merit.

Honors & Awards

  • ACS Petroleum Research Fund 25th-Anniversary Collection (2025) — first-author J. Phys. Chem. Lett. 2018 article selected as one of 40 highly cited PRF-funded articles published in the JPC journals since 2000.
  • ACS Most Read Physical Chemistry Articles of the Month (October 2017) — first-author J. Phys. Chem. Lett. 2017 article.
  • Top 10% of submissions, poster at the ACS Fall 2026 meeting, Chicago.
  • INSPIRE Fellowship, Department of Science and Technology, Government of India (2008–2012) — national fellowship for the top 1% of students pursuing science.
  • DAAD Summer Research Fellowship, Germany (2010) — one of 25 science students selected across India.
  • UKIERI Fellowship, UK–India Education and Research Initiative (2011) — summer research at the University of Surrey, UK.
  • Graduate student travel awards — Gordon Research Conference on Chemistry and Physics of Liquids (2015); American Conference on Theoretical Chemistry (2014).
  • Academic Excellence Award, IISER Pune (2010) — perfect-GPA semester.

Early Research Training & International Experience

National Chemical Laboratory, Pune (2008, 2009)
Hartree–Fock calculations of electronic properties of small molecules (Prof. Sourav Pal); Monte Carlo simulations of gas adsorption in metal–organic frameworks (Prof. Sudip Roy).
Saarland University, Saarbrücken, Germany (2010, DAAD)
Comparing the structure of nanoclusters (Prof. Michael Springborg).
University of Surrey, Guildford, UK (2011, UKIERI)
Electronic circular dichroism spectra of neuropeptides — a TD-DFT study (Prof. Brendan Howlin).
IISER Pune (2011–2012)
BS–MS thesis on polyethyleneimine–lipid bilayer interactions (Dr. Sudip Roy), published in Biomacromolecules 2013.

Working style

I enjoy building methods that are both physically grounded and computationally practical: clear assumptions, reproducible pipelines, and results that connect simulation to measurable observables.