GPCRmd uncovers the dynamics of the 3D-GPCRome

Rodríguez‐Espigares, Ismael; Torrens‐Fontanals, Mariona; Tiemann, Johanna K. S.; Aranda-García, David; Ramírez-Anguita, Juan Manuel; Stępniewski, Tomasz Maciej; Worp, Nathalie; Varela-Rial, Alejandro; Morales-Pastor, Adrián; Medel-Lacruz, Brian; Pándy-Szekeres, Gáspár; Mayol, Eduardo; Fonseca, Rasmus; Giorgino, Toni; Carlsson, Jens; Deupí, Xavier; Filipek, Sławomir; Gómez-Tamayo, José Carlos; González, Ángel F.; Gutiérrez‐de‐Terán, Hugo; Jimenez, Mireia; Jespers, Willem; Kapla, Jon; Kolb, Peter; Latek, Dorota; Martí-Solano, Maria; Matricon, P.; Matsoukas, Minos‐Timotheos; Miszta, Przemysław; Olivella, Mireia; Pérez‐Benito, Laura; Ríos, Santiago; Rodríguez-Torrecillas, Iván; Sallander, Jessica; Sztyler, Agnieszka; Vasile, Silvana; Hildebrand, Peter W.; Fabritiis, Gianni De; Gloriam, David E.; Cordomí, Arnau; Guixà-González, Ramón; Selent, Jana

doi:10.1101/839597

Cited by 16 publications

(26 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…and Ext. 9.42 ± 0.08 E-1 4.68 ± 0.00 E-1 Datasets Datasets used in this work are atomistic simulation trajectories of drug-target proteins reported in [19] and [20]. The two main datasets from [19] used are simulations of proteins in presence of FDA approved or underinvestigation molecules, as we aim to test the performance of ProGAE on capturing drug-induced structural variations.…”

Section: Training Loss S Protein Hace2mentioning

confidence: 99%

See 1 more Smart Citation

Learning Geometrically Disentangled Representations of Protein Folding Simulations

Tatro¹,

Das²,

Chen³

et al. 2022

Preprint

View full text Add to dashboard Cite

Massive molecular simulations of drug-target proteins have been used as a tool to understand disease mechanism and develop therapeutics. This work focuses on learning a generative neural network on a structural ensemble of a drug-target protein, e.g. SARS-CoV-2 Spike protein, obtained from computationally expensive molecular simulations. Model tasks involve characterizing the distinct structural fluctuations of the protein bound to various drug molecules, as well as efficient generation of protein conformations that can serve as an complement of a molecular simulation engine. Specifically, we present a geometric autoencoder framework to learn separate latent space encodings of the intrinsic and extrinsic geometries of the protein structure. For this purpose, the proposed Protein Geometric AutoEncoder (ProGAE) model is trained on the protein contact map and the orientation of the backbone bonds of the protein. Using ProGAE latent embeddings, we reconstruct and generate the conformational ensemble of a protein at or near the experimental resolution, while gaining better interpretability and controllability in term of protein structure generation from the learned latent space. Additionally, ProGAE models are transferable to a different state of the same protein or to a new protein of different size, where only the dense layer decoding from the latent representation needs to be retrained. Results show that our geometric learning-based method enjoys both accuracy and efficiency for generating complex structural variations, charting the path toward scalable and improved approaches for analyzing and enhancing high-cost simulations of drug-target proteins.

show abstract

Section: Training Loss S Protein Hace2mentioning

confidence: 99%

“…To show that a single ProGAE model can be trained on multiple different proteins, we also utilize simulations of 38 different G protein-coupled receptors from the GPCRmd dataset [20]. More information on these datasets is included in Appendix A.4.…”

Section: Training Loss S Protein Hace2mentioning

confidence: 99%

Learning Geometrically Disentangled Representations of Protein Folding Simulations

Tatro¹,

Das²,

Chen³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Finally, large collections of molecular dynamics (MD) trajectories [161,162] may be exploited toward protein flexibility learning. However, today unbiased MD simulations are still too short (and likely too inaccurate) to sample large conformational changes.…”

Section: Protein Disorder Flexibility and Dynamicsmentioning

confidence: 99%

Protein sequence‐to‐structure learning: Is this the end(‐to‐end revolution)?

et al. 2021

View full text Add to dashboard Cite

The potential of deep learning has been recognized in the protein structure prediction community for some time, and became indisputable after CASP13. In CASP14, deep learning has boosted the field to unanticipated levels reaching nearexperimental accuracy. This success comes from advances transferred from other machine learning areas, as well as methods specifically designed to deal with protein sequences and structures, and their abstractions. Novel emerging approaches include (i) geometric learning, that is, learning on representations such as graphs, threedimensional (3D) Voronoi tessellations, and point clouds; (ii) pretrained protein language models leveraging attention; (iii) equivariant architectures preserving the symmetry of 3D space; (iv) use of large meta-genome databases; (v) combinations of protein representations; and (vi) finally truly end-to-end architectures, that is, differentiable models starting from a sequence and returning a 3D structure. Here, we provide an overview and our opinion of the novel deep learning approaches developed in the last 2 years and widely used in CASP14.

show abstract

“…By these means, GPCRmd promotes transparency, consistency, and reproducibility in the field of GPCR dynamics and also facilitates data exchange between GPCR scientists from different disciplines. 24 Well before the official end of COST funding, the inclusiveness and influence of GLISTEN drove an intense community-wide desire to evolve a new COST Action. At the same time, developments in the field made clear that in order to meet current challenges, a wider perspective must be taken (explained in detail below).…”

Section: ■ Glisten and The Dawn Of Ernestmentioning

confidence: 99%

The European Research Network on Signal Transduction (ERNEST): Toward a Multidimensional Holistic Understanding of G Protein-Coupled Receptor Signaling

Sommer

Selent

Carlsson

et al. 2020

ACS Pharmacol. Transl. Sci.

Self Cite

View full text Add to dashboard Cite

G protein-coupled receptors (GPCRs) are intensively studied due to their therapeutic potential as drug targets. Members of this large family of transmembrane receptor proteins mediate signal transduction in diverse cell types and play key roles in human physiology and health. In 2013 the research consortium GLISTEN (COST Action CM1207) was founded with the goal of harnessing the substantial growth in knowledge of GPCR structure and dynamics to push forward the development of molecular modulators of GPCR function. The success of GLISTEN, coupled with new findings and paradigm shifts in the field, led in 2019 to the creation of a related consortium called ERNEST (COST Action CA18133). ERNEST broadens focus to entire signaling cascades, based on emerging ideas of how complexity and specificity in signal transduction are not determined by receptor–ligand interactions alone. A holistic approach that unites the diverse data and perspectives of the research community into a single multidimensional map holds great promise for improved drug design and therapeutic targeting.

show abstract

GPCRmd uncovers the dynamics of the 3D-GPCRome

Cited by 16 publications

References 54 publications

Learning Geometrically Disentangled Representations of Protein Folding Simulations

Learning Geometrically Disentangled Representations of Protein Folding Simulations

Protein sequence‐to‐structure learning: Is this the end(‐to‐end revolution)?

The European Research Network on Signal Transduction (ERNEST): Toward a Multidimensional Holistic Understanding of G Protein-Coupled Receptor Signaling

Contact Info

Product

Resources

About