EquiFold: Protein Structure Prediction with a Novel Coarse-Grained Structure Representation

Lee, Jaehyeon; Yadollahpour, Payman; Watkins, Andrew; Frey, Nathan C.; Leaver‐Fay, Andrew; Ra, Stephen; Cho, Kyunghyun; Gligorijević, Vladimir; Regev, Aviv; Bonneau, Richard

doi:10.1101/2022.10.07.511322

Cited by 24 publications

(29 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, DeepH3 was shown to outperform TrRosetta on antibodies [28, 29], while Nanonet obtains results of similar accuracy to AlphaFold2 on nanobodies with a far simpler architecture [30]. More recent examples of this are IgFold [25] and EquiFold [31], where the authors trained antibody-specific models that predict structures of comparable accuracy to AlphaFold-Multimer.…”

Section: Structure Predictionmentioning

confidence: 99%

ImmuneBuilder: Deep-Learning models for predicting the structures of immune proteins

Abanades

Wong

Boyles

et al. 2022

Preprint

View full text Add to dashboard Cite

Immune receptor proteins play a key role in the immune system and have shown great promise as biotherapeutics. The structure of these proteins is critical for understanding their antigen binding properties. Here, we present ImmuneBuilder, a set of deep learning models trained to accurately predict the structure of antibodies (ABodyBuilder2), nanobodies (NanoBodyBuilder2) and T-Cell receptors (TCRBuilder2). We show that ImmuneBuilder generates structures with state of the art accuracy while being far faster than AlphaFold2. For example, on a benchmark of 34 recently solved antibodies, ABodyBuilder2 predicts CDR-H3 loops with an RMSD of 2.81Å, a 0.09Å improvement over AlphaFold-Multimer, while being over a hundred times faster. Similar results are also achieved for nanobodies, (NanoBodyBuilder2 predicts CDR-H3 loops with an average RMSD of 2.89Å, a 0.55Å improvement over AlphaFold2) and TCRs. By predicting an ensemble of structures, ImmuneBuilder also gives an error estimate for every residue in its final prediction. ImmuneBuilder is made freely available, both to download (https://github.com/oxpig/ImmuneBuilder) and to use via our webserver (http://opig.stats.ox.ac.uk/webapps/newsabdab/sabpred). We also make available structural models for ~150 thousand non-redundant paired antibody sequences (https://zenodo.org/record/7258553).

show abstract

Section: Structure Predictionmentioning

confidence: 99%

ImmuneBuilder: Deep-Learning models for predicting the structures of immune proteins

Abanades

Wong

Boyles

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…We addressed the issue of generating structural antibody models of better physical quality without detriment to prediction of the backbone Cα. We trained a very simplistic deep learning model with a rudimentary loss function to minimize the risk of better peptide bond distances being the result of ingenious model architectures (Lee et al 2022; Ruffolo et al 2022; Abanades, Wong, et al 2022). We demonstrated that even such simple models can learn better quality peptide bonds if given the benefit of pre-training on a large augmented set of refined antibody structures.…”

Section: Discussionmentioning

confidence: 99%

“…To address this issue, but also to draw from AlphaFold2 and its derivatives (Ahdritz et al 2022; Lin et al 2022), many antibody-specific deep learning structure prediction methods have been introduced (Wilman et al 2022; Lin et al 2022). The algorithms started by addressing CDR-H3 loop prediction such as DeepH3 (Ruffolo et al 2020) and AbLooper (Abanades, Georges, et al 2022), after extending these to the entire variable domain via NanoNet (Cohen, Halfon, and Schneidman-Duhovny 2022) and the entire Fv molecule by DeepAb (Ruffolo, Sulam, and Gray 2022), IgFold (Ruffolo et al 2022), AbodyBuilder2 (Abanades, Wong, et al 2022), EquiFold (Lee et al 2022) and tFold-Ab (Wu et al 2022). As opposed to the homology methods that reported CDR-H3 root mean squared deviation (RMSD) accuracies in the region of 3-4Å (Almagro et al 2014), the deep learning methods achieve an RMSD of 2-3Å RMSD.…”

Section: Introductionmentioning

confidence: 99%

Structural pre-training improves physical accuracy of antibody structure prediction using deep learning

Kończak¹,

Janusz²,

Młokosiewicz³

et al. 2022

Preprint

View full text Add to dashboard Cite

Protein folding problem obtained a practical solution recently, owing to advances in deep learning. There are classes of proteins though, such as antibodies, that are structurally unique, where the general solution still lacks. In particular, the prediction of the CDR-H3 loop, which is an instrumental part of an antibody in its antigen recognition abilities, remains a challenge. Antibody-specific deep learning frameworks were proposed to tackle this problem noting great progress, both on accuracy and speed fronts. Oftentimes though, the original networks produce physically implausible bond geometries that then need to undergo a time-consuming energy minimization process. Here we hypothesized that pre-training the network on a large, augmented set of models with correct physical geometries, rather than a small set of real antibody X-ray structures, would allow the network to learn better bond geometries. We show that fine-tuning such a pre-trained network on a task of shape prediction on real X-ray structures improves the number of correct peptide bond distances. We further demonstrate that pre-training allows the network to produce physically plausible shapes on an artificial set of CDR-H3s, showing the ability to generalize to the vast antibody sequence space. We hope that our strategy will benefit the development of deep learning antibody models that rapidly generate physically plausible geometries, without the burden of time-consuming energy minimization.

show abstract

“…The database and web application TMvisDB offers a straightforward search functionality and visualization interface. While several accurate structure prediction methods have been made available over the last year [14][15][16][17], we chose to enhance TMvisDB sequence annotations with AlphaFold2 [12] predictions that have been shown to perform well in structural analysis of transmembrane proteins (TMPs) [52], and have, therefore, been successfully applied as input by resources such as the TmAlphaFold database that collects alpha-helical TMPs [34].…”

Section: Discussionmentioning

confidence: 99%

TMvisDB: resource for transmembrane protein annotation and 3D visualization

Marquet

Grekova

Houri

et al. 2022

Preprint

View full text Add to dashboard Cite

Since the rise of cellular organisms, transmembrane proteins (TMPs) are crucial to a variety of cellular processes due to their central role as gates and gatekeepers. Despite their importance, experimental high-resolution structures for TMPs remain underrepresented due to technical limitations. With structure prediction methods coming of age, predictions might fill some of the need. However, identifying the membrane regions and topology in three-dimensional structure files requires additional in silico prediction. Here, we introduce TMvisDB to sieve through millions of predicted structures for TMPs. This resource enables both, to browse through 46 million predicted TMPs and to visualize those along with their topological annotations. The database was created by joining AlphaFold DB structure predictions and transmembrane topology predictions from the protein language model based method TMbed. We show the utility of TMvisDB for individual proteins through two single use cases, namely the B-lymphocyte antigen CD20 (Homo sapiens) and the cellulose synthase (Novosphingobium sp. P6W). To demonstrate the value for large scale analyses, we focus on all TMPs predicted for the human proteome. TMvisDB is freely available at tmvis.predictprotein.org.

show abstract

EquiFold: Protein Structure Prediction with a Novel Coarse-Grained Structure Representation

Cited by 24 publications

References 32 publications

ImmuneBuilder: Deep-Learning models for predicting the structures of immune proteins

ImmuneBuilder: Deep-Learning models for predicting the structures of immune proteins

Structural pre-training improves physical accuracy of antibody structure prediction using deep learning

TMvisDB: resource for transmembrane protein annotation and 3D visualization

Contact Info

Product

Resources

About