prot_vec: Converting from protein sequences to vectors or vice versa.

In dongminjung/GenProSeq: Generating Protein Sequences with Deep Generative Models

Converting from protein sequences to vectors or vice versa.

Description

By using the word2vec model, amino acids are mapped to vectors of real numbers. Conceptually, it involves a mathematical embedding from a space with many dimensions per amino acid to a continuous vector space with a much lower dimension.

Usage

prot2vec(prot_seq, embedding_dim, embedding_matrix = NULL, ...)
vec2prot(prot_vec, embedding_matrix)

Arguments

prot_seq	protein sequences
prot_vec	protein embedding vectors
embedding_dim	dimension of embedding vectors
embedding_matrix	embedding matrix (default: NULL)
...	arguments for "word2vec::word2vec" but for dim, min_count and split

Value

prot_seq	protein sequences
prot_vec	protein embedding vectors
embedding_matrix	embedding matrix

Author(s)

Dongmin Jung

References

Mikolov, T., Sutskever, I., Chen, K., Corrado, G., & Dean, J. (2013). Distributed Representations of Words and Phrases and their Compositionality. arXiv:1310.4546.

Chang, M. (2020). Artificial intelligence for drug development, precision medicine, and healthcare.

See Also

word2vec::word2vec, word2vec::word2vec_similarity

Examples

prot_seq <- example_PTEN[1:10]
prot2vec_result <- prot2vec(prot_seq = prot_seq, embedding_dim = 8)
vec2prot_result <- vec2prot(prot_vec = prot2vec_result$prot_vec,
                            embedding_matrix = prot2vec_result$embedding_matrix)

dongminjung/GenProSeq documentation built on May 3, 2022, 10:28 p.m.