Search for: All records

The HIV-1 capsid protein (CA) assembles into a conical shell during viral maturation, encasing and protecting the viral RNA genome. The C-terminal domain (CTD) of the two-domain capsid protein dimerizes, and this dimer connects individual chains in the mature capsid lattice. Previous NMR studies have shown that different dimer arrangements can be formed by isolated capsid protein chains and in assembled capsid lattices; however, the dynamics and functional relevance of these alternate dimers are unknown. To explore the conformational landscape of the CA-CTD dimer, we carried out atomistic molecular dynamics simulations using the weighted ensemble path sampling strategy, generating an ensemble of conformations. Focusing on the two dimer forms previously observed via solution NMR, we refined the conformational ensemble to highlight two metastable states using a Markov state model. Experimentally, we measured the interconversion rates between the two alternate dimers using¹⁹F NMR, and these rates showed good agreement with the interconversion rates derived from the simulations. After identifying the key interactions that distinguish the dimer states, the alternate dimer was further experimentally verified through disulfide crosslinking. Our results demonstrate the advantages of pairing weighted ensemble path sampling with¹⁹F NMR to gain atomistic insights into the hidden dimer state of the HIV-1 capsid protein. 

Sequence-encoded folding is the foundation of protein structure and is also possible in synthetic chains of artificial chemical composition. In natural proteins, the characteristics of the unfolded state are as important as those of the folded state in determining folding energetics. While much is known about folded structures adopted by artificial protein-like chains, corresponding information about the unfolded states of these molecules is lacking. Here, we report the consequences of altered backbone composition on the structure, stability, and dynamics of the folded and unfolded states of a compact helix-rich protein. Characterization through a combination of biophysical experiments and atomistic simulation reveals effects of backbone modification that depend on both the type of artificial monomers employed and where they are applied in sequence. In general, introducing artificial connectivity in a way that reinforces characteristics of the unfolded state ensemble of the prototype natural protein minimizes the impact of chemical changes on folded stability. These findings have implications in the design of protein mimetics and provide an atomically detailed picture of the unfolded state of a natural protein and artificial analogues under non-denaturing conditions. 

In the Big Data era, a change of paradigm in the use of molecular dynamics is required. Trajectories should be stored under FAIR (findable, accessible, interoperable and reusable) requirements to favor its reuse by the community under an open science paradigm. 

We develop a generalizable AI-driven workflow that leverages heterogeneous HPC resources to explore the time-dependent dynamics of molecular systems. We use this workflow to investigate the mechanisms of infectivity of the SARS-CoV-2 spike protein, the main viral infection machinery. Our workflow enables more efficient investigation of spike dynamics in a variety of complex environments, including within a complete SARS-CoV-2 viral envelope simulation, which contains 305 million atoms and shows strong scaling on ORNL Summit using NAMD. We present several novel scientific discoveries, including the elucidation of the spike’s full glycan shield, the role of spike glycans in modulating the infectivity of the virus, and the characterization of the flexible interactions between the spike and the human ACE2 receptor. We also demonstrate how AI can accelerate conformational sampling across different systems and pave the way for the future application of such methods to additional studies in SARS-CoV-2 and other molecular systems.