Biology subjects

Salike, S.

Publications and source records attributed to Salike, S..

6 recordsLinked to original sources

bioRxiv2026

Reliable and Efficient Agentic Protein Design with Toolset Harnessing

Computational protein design workflows require the orchestration of multiple tools, data formats, and analysis steps that are often fragmented. Here, we introduce Chaperone, a lightweight and customizable framework that enables seamless integration of RFdiffusion3 (RFD3) with downstream sequence design, structure prediction, and evaluation tools, including ProteinMPNN, AlphaFold3 (AF3), and more than 50 other protein design tools. By combining a programmable agent-friendly interface with parallelized execution on high-performance computing systems, Chaperone as a harness enables rapid iteration and scalable exploration of complex protein design workflows. Across an evaluation of nine biomolecular design tasks, we find using Chaperone reduces total token usage relative to baseline agents by 86% and GPU costs by 64%, while increasing the overall computational design pass rate. We demonstrate Chaperones utility for large-scale design-space exploration through a two-week end-to-end small-molecule binder design campaign, yielding low nanomolar binders to cortisol zero shot. We further demonstrate its versatility across distinct design challenges, including the design of a symmetric 12-subunit self-assembly nanoparticle and a linker that mediates interactions between C2- and D3-symmetric protein complexes to form crystalline assemblies. Together, these results demonstrate how agent harnesses built around standardized tool interfaces can substantially improve the efficiency and success of complex, long-horizon agentic design workflows in biology. We make Chaperone freely available to the community on GitHub: https://github.com/RosettaCommons/chaperone.

biochemistry↗

bioRxiv2026

De novo design of allosterically controlled enzymes

The ability of enzymes to sense signals and respond by changing their structure and activity underlies cellular processes from signaling to metabolic control. While there have been recent advances in the de novo design of enzymes and conformationally switching proteins, combining these to achieve allosteric regulation of enzymatic activity in a fully designed system remains an outstanding challenge. Here, we show that denoising diffusion models enable the design of compact de novo enzymes whose activity can be allosterically activated or repressed by designed protein effectors. Our results establish a general route to allosteric control over a wide range of catalytic and other protein functions.

biochemistry↗

bioRxiv2025

De novo Design of All-atom Biomolecular Interactions with RFdiffusion3

Deep learning has accelerated protein design, but most existing methods are restricted to generating protein backbone coordinates and often neglect interactions with other biomolecules. We present RFdiffusion3 (RFD3), a diffusion model that generates protein structures in the context of ligands, nucleic acids and other non-protein constellations of atoms. Because all polymer atoms are modeled explicitly, conditioning the model on complex sets of atom-level constraints for enzyme design and other challenges is both simpler and more effective than previous approaches. RFD3 achieves improved performance compared to prior approaches on a range of in silico benchmarks with one tenth the computational cost. Finally, we demonstrate the broad applicability of RFD3 by designing and experimentally characterizing DNA binding proteins and cysteine hydrolases. The ability to rapidly generate protein structures guided by complex sets of atom-level constraints in the context of arbitrary non-protein atoms should further expand the range of functions attainable through protein design.

biochemistry↗

bioRxiv2025

Atom level enzyme active site scaffolding using RFdiffusion2

De novo enzyme design starts from ideal active site descriptions consisting of constellations of catalytic residue functional groups around reaction transition state(s), and seeks to generate protein structures that can accurately hold the site in place. Highly active enzymes have been designed starting from such descriptions using the generative AI method RFdiffusion [1-3], but there are two current methodological limitations. First, the geometry of the active site can only be specified at the residue level, so for each catalytic residue functional group placed around the reaction transition state, the possible locations of the residue backbone must be enumerated by building side chain rotamers back from the functional group. Second, the location of the catalytic residues along the sequence must be specified in advance, which considerably limits the space of solutions which can be sampled. Here we describe a new deep generative method, Rosetta Fold diffusion 2 (RFdiffusion2), that solves both problems, enabling enzymes to be designed from sequence agnostic descriptions of functional group locations without inverse rotamer generation. We first evaluate RFdiffusion2 on an in silico enzyme design benchmark of 41 diverse active sites and find that it is able to successfully build proteins scaffolding all 41 sites, compared to 16/41 with prior state-of-the-art deep learning methods. Next, we design enzymes around three diverse catalytic sites and characterize the designs experimentally; in each case we identify active catalysts in testing less than 96 sequences. RFdiffusion2 demonstrates the potential of atomic resolution generative models for the design of de novo enzymes directly from their reaction mechanisms.

biochemistry↗

bioRxiv2024

Computational Design of Metallohydrolases

De novo enzyme design starts from a description of an ideal active site composed of catalytic residues surrounding the reaction transition state(s), and builds a protein structure that contains this site1-7. Generative AI methods such as RFdiffusion11,12 now enable the direct generation of proteins around active sites, but to date, such scaffolding has required specification of both the position in the sequence and the backbone coordinates of each catalytic residue, which complicates sampling. Here we introduce a generative AI method called RFdiffusion2 that overcomes these limitations and use it to design zinc metallohydrolases starting from a density functional theory description of the active site geometry. Of an initial set of 96 designs tested experimentally, the most active has a kcat/KM of 16,000 M-1 s-1, orders of magnitude higher than previously designed metallohydrolases.6,7,13,14 A second round of 96 designs yielded 3 additional highly active enzymes, with kcat/KM up to 53,000 M-1 s-1 and kcat up to 1.5 s-1. The structures of the four enzymes are very different from each other and from the structures in the PDB. Each enzyme positions the reaction substrate almost perfectly for nucleophilic attack by a water molecule activated by the bound metal, and are predicted by PLACER15 and Chai-144 to have highly preorganized active sites. The ability to generate highly active catalysts straight out of the computer, without experimental optimization, using quantum chemistry calculated active site geometries should open the door to a new generation of potent designer enzymes.16,17

biochemistry↗

bioRxiv2018

Thermodynamically Consistent Estimation of Gibbs Free Energy from Data: Data Reconciliation Approach

MotivationThermodynamic analysis of biological reaction networks requires the availability of accurate and consistent values of Gibbs free energies of reaction and formation. These Gibbs energies can be measured directly via the careful design of experiments or can be computed from the curated Gibbs free energy databases. However, the computed Gibbs free energies of reactions and formations do not satisfy the thermodynamic constraints due to the compounding effect of measurement errors in the experimental data. The propagation of these errors can lead to a false prediction of pathway feasibility and uncertainty in the estimation of thermodynamic parameters.\n\nResultsThis work proposes a data reconciliation framework for thermodynamically consistent estimation of Gibbs free energies of reaction, formation and group contributions from experimental data. In this framework, we formulate constrained optimization problems that reduce measurement errors and their effects on the estimation of Gibbs energies such that the thermodynamic constraints are satisfied. When a subset of Gibbs free energies of formations is unavailable, it is shown that the accuracy of their resulting estimates is better than that of existing empirical prediction methods. Moreover, we also show that the estimation of group contributions can be improved using this approach. Further, we provide guidelines based on this approach for performing systematic experiments to estimate unknown Gibbs formation energies.\n\nAvailabilityThe MATLAB code for the executing the proposed algorithm is available for free on the GitHub repository: https://github.com/samansalike/DR-thermo\n\nContactniravbhatt@iitm.ac.in

systems biology↗