2026-10-15T01:04:07Z https://www.scidb.cn/oai

10.57760/sciencedb.011fm 2026-09-13T19:34:24Z

2026-09-13 AWP-Fusion_Project doi:10.57760/sciencedb.011fm en This dataset contains ESM-2 protein language model feature vectors (embeddings) for three enzyme property prediction tasks: optimal pH (pHopt), optimal temperature (Topt), and melting temperature (Tm). The features were generated using the ESM-2 (650M parameters) model for each protein sequence. These embeddings serve as input representations for the AWP-Fusion model, a lightweight adaptive weighted pooling fusion model proposed for predicting enzyme properties from protein sequences. The dataset supports the reproducibility of the experimental results reported in the associated manuscript. ESM-2 embeddings; protein language model; enzyme property prediction Wei Wang Liyan Shen Feng Wang Shan Chang PUBLIC https://creativecommons.org/licenses/by/4.0/ dataset Science Data Bank