An official website of the United States government
Abstract We introduceBoltzGen, an all-atom generative model for designing proteins and peptides across all modalities to bind a wide range of biomolecular targets. BoltzGen builds strong structural reasoning capabilities about target-binder interactions into its generative design process. This is achieved by unifying design and structure prediction, resulting in a single model that also reaches state-of-the-art folding performance. BoltzGen’s generation process can be controlled with a flexible design specification language over covalent bonds, structure constraints, binding sites, and more. We experimentally validate these capabilities in a total of eight diverse wetlab design campaigns with functional and affinity readouts across 26 targets. The experiments span binder modalities from nanobodies to disulfide-bonded peptides and include targets ranging from disordered proteins to small molecules. For instance, we test 15 nanobody and protein binder designs against each of nine novel targets with low similarity to any protein with a known bound structure. For both binder modalities, this yields nanomolar binders for 66% of targets. We release model weights, data, and both inference and training code at:https://github.com/HannesStark/boltzgen.
more »
« less
BinderSpace: A package for sequence space analyses for datasets of affinity‐selected oligonucleotides and peptide‐based molecules
Abstract Discovery of target‐binding molecules, such as aptamers and peptides, is usually performed with the use of high‐throughput experimental screening methods. These methods typically generate large datasets of sequences of target‐binding molecules, which can be enriched with high affinity binders. However, the identification of the highest affinity binders from these large datasets often requires additional low‐throughput experiments or other approaches. Bioinformatics‐based analyses could be helpful to better understand these large datasets and identify the parts of the sequence space enriched with high affinity binders.BinderSpaceis an open‐source Python package that performs motif analysis, sequence space visualization, clustering analyses, and sequence extraction from clusters of interest. The motif analysis, resulting in text‐based and visual output of motifs, can also provide heat maps of previously measured user‐defined functional properties for all the motif‐containing molecules. Users can also run principal component analysis (PCA) and t‐distributed stochastic neighbor embedding (t‐SNE) analyses on whole datasets and on motif‐related subsets of the data. Functionally important sequences can also be highlighted in the resulting PCA and t‐SNE maps. If points (sequences) in two‐dimensional maps in PCA or t‐SNE space form clusters, users can perform clustering analyses on their data, and extract sequences from clusters of interest. We demonstrate the use ofBinderSpaceon a dataset of oligonucleotides binding to single‐wall carbon nanotubes in the presence and absence of a bioanalyte, and on a dataset of cyclic peptidomimetics binding to bovine carbonic anhydrase protein.BinderSpaceis openly accessible to the public via the GitHub website:https://github.com/vukoviclab/BinderSpace.
more »
« less
- Award ID(s):
- 2106587
- PAR ID:
- 10505124
- Publisher / Repository:
- J. Comput. Chem.
- Date Published:
- Journal Name:
- Journal of Computational Chemistry
- Volume:
- 44
- Issue:
- 22
- ISSN:
- 0192-8651
- Page Range / eLocation ID:
- 1836 to 1844
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Clustering is a fundamental task in machine learning. One of the most successful and broadly used algorithms is DBSCAN, a density-based clustering algorithm. DBSCAN requires ϵ-nearest neighbor graphs of the input dataset, which are computed with range-search algorithms and spatial data structures like KD-trees. Despite many efforts to design scalable implementations for DBSCAN, existing work is limited to low-dimensional datasets, as constructing ϵ-nearest neighbor graphs can be expensive in high-dimensions. This article introduces a modified DBSCAN, usingk-nearest neighbor (kNN) graphs to improve efficiency. We outline conditions forkNN-DBSCAN to match DBSCAN’s results and present a parallel implementation using OpenMP and MPI for shared and distributed memory systems. Testing on datasets up to 32 dimensions, we achieve remarkable scalability. Our implementation clusters one billion 3D points in under one second on 28K cores at TACC’s Frontera system. In a larger run, we cluster 65 billion points in 20 dimensions in under 40 seconds using 114,688 cores. Our method is up to 37× faster than state-of-the-art parallel DBSCAN on a 20-dimensional dataset with 4 million points. Code is available athttps://github.com/ut-padas/knndbscan.more » « less
-
Connectomics, a subfield of neuroscience, reconstructs structural and functional brain maps at synapse-level resolution. These complex spatial maps consist of tree-like neurons interconnected by synapses. Motif analysis is a widely used method for identifying recurring subgraph patterns in connectomes. These motifs, thus, potentially represent fundamental units of information processing. However, existing computational tools often oversimplify neurons as mere nodes in a graph, disregarding their intricate morphologies. In this paper, we introduceMoMo, a novel interactive visualization framework for analyzingneuron morphology-aware motifsin large connectome graphs. First, we propose an advanced graph data structure that integrates both neuronal morphology and synaptic connectivity. This enables highly efficient, parallel subgraph isomorphism searches, allowing for interactive morphological motif queries. Second, we develop a sketch-based interface that facilitates the intuitive exploration of morphology-based motifs within our new data structure. Users can conduct interactive motif searches on state-of-the-art connectomes and visualize results as interactive 3D renderings. We present a detailed goal and task analysis for motif exploration in connectomes, incorporating neuron morphology. Finally, we evaluateMoMothrough case studies with four domain experts, who asses the tool’s usefulness and effectiveness in motif exploration, and relevance to real-world neuroscience research. The source code forMoMois availablehere.more » « less
-
Abstract Protein language models, like the popular ESM2, are widely used tools for extracting evolution-based protein representations and have achieved significant success on downstream biological tasks. Representations based on sequence and structure models, however, show significant performance differences depending on the downstream task. A major open problem is to obtain representations that best capture both the evolutionary and structural properties of proteins in general. Here we introduceImplicitStructureModel(ISM), a sequence-only input model with structurally-enriched representations that outperforms state-of-the-art sequence models on several well-studied benchmarks including mutation stability assessment and structure prediction. Our key innovations are a microenvironment-based autoencoder for generating structure tokens and a self-supervised training objective that distills these tokens into ESM2’s pre-trained model. We have madeISM’s structure-enriched weights easily available: integrating ISM into any application using ESM2 requires changing only a single line of code. Our code is available athttps://github.com/jozhang97/ISM.more » « less
-
Abstract Pangenomes are becoming increasingly popular data structures for genomics analyses due to their ability to compactly represent the genetic diversity within populations. Constructing a pangenome graph, however, is still a time-consuming and expensive process. A promising approach for pangenome construction consists of progressively augmenting a pangenome graph with additional high-quality assemblies. Currently, there is no method for augmenting a pangenome graph with unassembled reads from newly sequenced samples without first aligning the reads to a reference genome and performing variant calling and genotyping on the new individuals. In this work, we present the first assembly-free and mapping-free approach for augmenting an existing pangenome graph using unassembled long reads from an individual not already present in the pangenome. Our approach consists of finding sample specific sequences in reads using efficient indexes, clustering reads corresponding to the same novel variant(s), and then building a consensus sequence to be added to the pangenome graph for each variant separately. Using simulated reads based on Human Pangenome Reference Consortium (HPRC) assemblies, we demonstrate the effectiveness of the proposed approach for progressively augmenting the pangenome with long reads, without the need forde novoassembly or predicting genetic variants of the new sample. The software is freely available athttps://github.com/ldenti/palss.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript
- Journal Article:
- https://doi.org/10.1002/jcc.27130
