Efficient generative modeling of protein sequences using simple autoregressive models.

Autor: Trinquier J; Sorbonne Université, CNRS, Institut de Biologie Paris Seine, Biologie Computationnelle et Quantitative LCQB, F-75005, Paris, France.; Laboratoire de Physique de l'Ecole Normale Supérieure, ENS, Université PSL, CNRS, Sorbonne Université, Université de Paris, F-75005, Paris, France., Uguzzoni G; Department of Applied Science and Technology (DISAT), Politecnico di Torino, Corso Duca degli Abruzzi 24, I-10129, Torino, Italy.; Italian Institute for Genomic Medicine, IRCCS Candiolo, SP-142, I-10060, Candiolo (TO), Italy., Pagnani A; Department of Applied Science and Technology (DISAT), Politecnico di Torino, Corso Duca degli Abruzzi 24, I-10129, Torino, Italy.; Italian Institute for Genomic Medicine, IRCCS Candiolo, SP-142, I-10060, Candiolo (TO), Italy.; INFN Sezione di Torino, Via P. Giuria 1, I-10125, Torino, Italy., Zamponi F; Laboratoire de Physique de l'Ecole Normale Supérieure, ENS, Université PSL, CNRS, Sorbonne Université, Université de Paris, F-75005, Paris, France., Weigt M; Sorbonne Université, CNRS, Institut de Biologie Paris Seine, Biologie Computationnelle et Quantitative LCQB, F-75005, Paris, France. martin.weigt@sorbonne-universite.fr.
Jazyk: angličtina
Zdroj: Nature communications [Nat Commun] 2021 Oct 04; Vol. 12 (1), pp. 5800. Date of Electronic Publication: 2021 Oct 04.
DOI: 10.1038/s41467-021-25756-4
Abstrakt: Generative models emerge as promising candidates for novel sequence-data driven approaches to protein design, and for the extraction of structural and functional information about proteins deeply hidden in rapidly growing sequence databases. Here we propose simple autoregressive models as highly accurate but computationally efficient generative sequence models. We show that they perform similarly to existing approaches based on Boltzmann machines or deep generative models, but at a substantially lower computational cost (by a factor between 10 2 and 10 3 ). Furthermore, the simple structure of our models has distinctive mathematical advantages, which translate into an improved applicability in sequence generation and evaluation. Within these models, we can easily estimate both the probability of a given sequence, and, using the model's entropy, the size of the functional sequence space related to a specific protein family. In the example of response regulators, we find a huge number of ca. 10 68 possible sequences, which nevertheless constitute only the astronomically small fraction 10 -80 of all amino-acid sequences of the same length. These findings illustrate the potential and the difficulty in exploring sequence space via generative sequence models.
(© 2021. The Author(s).)
Databáze: MEDLINE
Externí odkaz: