LA4SR: illuminating the dark proteome with generative AI

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Nelson, David R., Jaiswal, Ashish Kumar, Ismail, Noha, Mystikou, Alexandra, Salehi-Ashtiani, Kourosh
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866910738815123456
author Nelson, David R.
Jaiswal, Ashish Kumar
Ismail, Noha
Mystikou, Alexandra
Salehi-Ashtiani, Kourosh
author_facet Nelson, David R.
Jaiswal, Ashish Kumar
Ismail, Noha
Mystikou, Alexandra
Salehi-Ashtiani, Kourosh
contents AI language models (LMs) show promise for biological sequence analysis. We re-engineered open-source LMs (GPT-2, BLOOM, DistilRoBERTa, ELECTRA, and Mamba, ranging from 70M to 12B parameters) for microbial sequence classification. The models achieved F1 scores up to 95 and operated 16,580x faster and at 2.9x the recall of BLASTP. They effectively classified the algal dark proteome - uncharacterized proteins comprising about 65% of total proteins - validated on new data including a new, complete Hi-C/Pacbio Chlamydomonas genome. Larger (>1B) LA4SR models reached high accuracy (F1 > 86) when trained on less than 2% of available data, rapidly achieving strong generalization capacity. High accuracy was achieved when training data had intact or scrambled terminal information, demonstrating robust generalization to incomplete sequences. Finally, we provide custom AI explainability software tools for attributing amino acid patterns to AI generative processes and interpret their outputs in evolutionary and biophysical contexts.
format Preprint
id arxiv_https___arxiv_org_abs_2411_06798
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle LA4SR: illuminating the dark proteome with generative AI
Nelson, David R.
Jaiswal, Ashish Kumar
Ismail, Noha
Mystikou, Alexandra
Salehi-Ashtiani, Kourosh
Genomics
Artificial Intelligence
Computation and Language
Quantitative Methods
AI language models (LMs) show promise for biological sequence analysis. We re-engineered open-source LMs (GPT-2, BLOOM, DistilRoBERTa, ELECTRA, and Mamba, ranging from 70M to 12B parameters) for microbial sequence classification. The models achieved F1 scores up to 95 and operated 16,580x faster and at 2.9x the recall of BLASTP. They effectively classified the algal dark proteome - uncharacterized proteins comprising about 65% of total proteins - validated on new data including a new, complete Hi-C/Pacbio Chlamydomonas genome. Larger (>1B) LA4SR models reached high accuracy (F1 > 86) when trained on less than 2% of available data, rapidly achieving strong generalization capacity. High accuracy was achieved when training data had intact or scrambled terminal information, demonstrating robust generalization to incomplete sequences. Finally, we provide custom AI explainability software tools for attributing amino acid patterns to AI generative processes and interpret their outputs in evolutionary and biophysical contexts.
title LA4SR: illuminating the dark proteome with generative AI
topic Genomics
Artificial Intelligence
Computation and Language
Quantitative Methods
url https://arxiv.org/abs/2411.06798