Inclusion of phylogenetic information affects cost, accuracy, and generalizability

Association·arcadia

Claim that explicit inclusion of phylogenetic information into protein models can reduce training cost, improve model accuracy, and expand generalizability.

Confidence

80%

active

Evidence Quote

“explicit inclusion of phylogenetic information ... may reduce training cost, improve model accuracy, and expand generalizability”

Arguments

Explicit phylogenetic information in protein modelssubject

Training cost of protein modelsobject

Connections (4)

Model accuracyFactor

Generalizability of protein modelsFactor

How phylogenetic data engineering expands protein model generalizabilityInferenceChain

Petabase-scale sequence alignment increases viral discoveryAssociation

Evidence

“Evidence summarizing advances in de novo protein design, emphasizing progress from new protein structures to programmable functions as described by Kortemme (2024).”

Kortemme T. (2024). De novo protein design—From new structures to programmable functions doi:10.1016/j.cell.2023.12.028 ↗

“Evidence highlighting the maturation of de novo protein design, summarizing progress and challenges in designing proteins from first principles, as discussed in Nature (2016).”

(2016). The coming of age of de novo protein design doi:10.1038/nature19946 ↗