From 8c1b0b1c51353ac488acf6e202cb0e3b1dd65120 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Konstantin=20Wei=C3=9Fenow?= <60698240+kWeissenow@users.noreply.github.com> Date: Thu, 14 Jul 2022 23:10:15 +0200 Subject: [PATCH] Update README.md --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index 19bbfdb..825523e 100644 --- a/README.md +++ b/README.md @@ -46,7 +46,7 @@ You can compute structure predictions based on FASTA sequences using python predict.py -i -o ``` -The ProtT5 language model will be downloaded on first use and stored by default in the directory 'ProtT5-XL-U50'. You can change this directory with the `--t5_model` parameter. +The ProtT5 protein language model used to generate sequence embeddings will be downloaded on first use (~7 GB) and stored by default in the directory 'ProtT5-XL-U50'. You can change this directory with the `--t5_model` parameter. By default, the script will produce PDB files and distance maps. You can disable outputs using the parameters `--no-pdb` and `--no-distance-maps` respectively. ## Mutation effect prediction