August 7, 2025PLoS ONEOpen Access

Bag-of-words is competitive with sum-of-embeddings language-inspired representations on protein inference

View Full Paper

Ask AI

Bookmark

Authors

Discussion

Analysis shows bag-of-words outperforms self-supervised learning for protein inference, indicating potential in feature selection.

Main finding reveals that bag-of-words representations are more effective than those based on self-supervised learning for protein inference tasks.
Key evidence shows simple bag-of-words histograms capture important discriminant features for data-driven function prediction.
Approach involved testing protein sequence representations on a large protein database across multiple inference tasks.
Significance lies in the potential revisions to self-supervised learning strategies, prompting exploration of alternative schemes for protein representation.

Papadopoulos et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark

August 7, 2025PLoS ONEOpen Access

View Full Paper

Ask AI

Bookmark

Authors

Analysis shows bag-of-words outperforms self-supervised learning for protein inference, indicating potential in feature selection.

Main finding reveals that bag-of-words representations are more effective than those based on self-supervised learning for protein inference tasks.
Key evidence shows simple bag-of-words histograms capture important discriminant features for data-driven function prediction.
Approach involved testing protein sequence representations on a large protein database across multiple inference tasks.
Significance lies in the potential revisions to self-supervised learning strategies, prompting exploration of alternative schemes for protein representation.

Papadopoulos et al. (2025) studied this question.

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

View Full Paper

Ask AI

Bookmark