Figure - PMC

Skip to main content

An official website of the United States government

Here's how you know

Here's how you know

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

View full-text article in PMC

. 2021 Nov 19;1(1):vbab035. doi: 10.1093/bioadv/vbab035

Search in PMC
Search in PubMed
View in NLM Catalog
Add to search

© The Author(s) 2021. Published by Oxford University Press.

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.

PMC Copyright notice

Fig. 4. — Qualitative analysis confirmed LA to be effective. UMAP (McInnes et al., 2018) projections of per-protein embeddings colored according to subcellular location (setDeepLoc). Both plots were created with the same default values of the python umap-learn library. Top: ProtT5 embeddings (LA input; x) mean-pooled over protein length (as for FNN/EAT input). Bottom: ProtT5 embeddings (LA input; x) weighted according to the attention distribution produced by LA (this is not $x^{'}$ as we sum the input features x and not the values v after the convolution)