Embeddings from protein language models predict conservation and variant effects
Article
[키워드] Algorithm
alignment
Amino acid
analyses
applied
approach
data set
database
deep
deep learning
Effect
embedding
entire sequence
expand
experiment
fraction
human proteins
investigated
language
Local
Logistic regression
magnitude
Model
Multiple sequence alignment
Nvidia
Pearson correlation
predict
predicted
prediction
Probability
Protein
protein sequences
Proteins
proteome
regions
residue
RTX
SARS-CoV-2 variant
scanning
score
sequence
Sequence conservation
significantly
Spearman
statistically
supplementary material
variant
while
[DOI] 10.1007/s00439-021-02411-y PMC 바로가기
[DOI] 10.1007/s00439-021-02411-y PMC 바로가기