Mirrors_Framework/deepmind-research
1
0
Fork 0
mirror of https://github.com/google-deepmind/deepmind-research.git synced 2026-05-10 05:17:46 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

Fix the README: We are not using mutual_information in the model.

Browse Source 
PiperOrigin-RevId: 305474614
This commit is contained in:
Augustin Zidek
2020-04-08 15:43:03 +01:00
committed by Diego de Las Casas
parent 47d3f0fc6e
commit 25667f7d79
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
alphafold_casp13/README.md
+1 -1
View File
@@ -200,7 +200,7 @@ thorough explanation are explained in the section below the table. Note that
| `hhblits_profile` | ❌ | float32 | `(NR, 22)` | A profile (probability distribution over amino acid types) computed using HHBlits MSA. Encoding: 20 amino acids + 'X' + '-'. |
| `hmm_profile` | ✔️ | float32 | `(NR, 30)` | The HHBlits HHM profile (from the `-ohhm` HHBlits output file). Asterisks in the output are replaced by 0.0. See below. |
| `key` | ❌ | string | `(1)` | The unique id of the protein. |
| `mutual_information` | ✔️ | float32 | `(NR, NR, 1)` | The average product corrected mutual information. See https://doi.org/10.1093/bioinformatics/btm604. |
| `mutual_information` | ❌ | float32 | `(NR, NR, 1)` | The average product corrected mutual information. See https://doi.org/10.1093/bioinformatics/btm604. |
| `non_gapped_profile` | ✔️ | float32 | `(NR, 21)` | A profile from amino acids only (discounting gaps). See below. |
| `num_alignments` | ✔️ | int64 | `(NR, 1)` | The number of HHBlits multiple sequence alignments. Has to be repeated `NR` times. See below. |
| `num_effective_alignments` | ❌ | float32 | `(1)` | The number of effective alignments (neff at 62 % sequence similarity). |