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.
Author:Jeanne Trinquier, Jeanne Trinquier, Guido Uguzzoni, Andrea Pagnani, Francesco Zamponi, Martin WeigtPublished:2021
Here we propose simpleautoregressivemodels as highly accurate but computationally efficientgenerativesequencemodels. We show that they perform similarly to existing approaches based on Boltzmann machines or deep generativemodels, but at a substantially lower computational cost (by a factor between 10 2 and 10 3). Furthermore, the simple ...
Generativemodels 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 simpleautoregressivemodels as highly accurate but computationally efficientgenerativesequencemodels. We show that they perform ...
Author:Jeanne Trinquier, Jeanne Trinquier, Guido Uguzzoni, Andrea Pagnani, Francesco Zamponi, Martin WeigtPublished:2021
our autoregressivemodels are the only generativemod-els we know about, which allow for calculating exact sequence probabilities, and not only non-normalized se-quence weights, thereby enabling the comparison of the same sequence in di erent models for di erent protein families. II. RESULTS A. Autoregressivemodels for protein families
Author:Jeanne Trinquier, Jeanne Trinquier, Guido Uguzzoni, Andrea Pagnani, Francesco Zamponi, Martin WeigtPublished:2021
Here we propose simpleautoregressivemodels as highly accurate but computationally efficientgenerativesequencemodels. We show that they perform similarly to existing approaches based on Boltzmann machines or deep generativemodels, but at a substantially lower computational cost (by a factor between 102 and 103).
Boltzmann machines or deep generativemodels, but at a substantially lower computational cost (by a factor between 102 and 103). 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 ...
Simpleautoregressivemodels are proposed as highly accurate but computationally efficientgenerativesequencemodels that perform similarly to existing approaches based on Boltzmann machines or deep generativemodels, but at a substantially lower computational cost. Generativemodels emerge as promising candidates for novel sequence-data driven approaches to protein design, and for the ...
Generativemodels emerge as promising candidates for novel sequence-data driven approaches to protein design, and for the extraction of structural and functional information about
Generativemodels 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 simpleautoregressivemodels as highly accurate but computationally extremely efficientgenerativesequencemodels. We show that they ...