Gemma I Mtnez (IBE, CSIC) present the tool FANTASIA
Автор: Conexión CSIC BCB
Загружено: 2025-06-06
Просмотров: 53
The function of many protein-coding genes remains poorly characterized, especially in non-model organisms. Traditional sequence homology-based methods often fall short in accurately transferring functional annotations. With the rapid sequencing and release of genomes from non-model organisms, there is a growing need for faster and more scalable sequence-based functional prediction methods. FANTASIA (Functional ANnoTAtion based on embedding space SImilArity) is a pipeline designed for annotating Gene Ontology (GO) terms for protein sequences using advanced protein language models such as ProtT5, ProstT5, and ESM2. It accepts a proteome file as input (either the longest isoform or the full set of isoforms for all genes), preprocesses the sequences, and converts them into embeddings. These embeddings are then analyzed with GOPredSim, which assigns functional annotations to nearly all genes in a proteome based on their embedding similarity to an annotated reference. FANTASIA provides a scalable, efficient solution for protein functionality analysis, overcoming the limitations of traditional sequence homology-based GO annotation. The current version of the tool is freely available as an open-access Singularity container.
Доступные форматы для скачивания:
Скачать видео mp4
-
Информация по загрузке:
