Update README.md
Browse files
README.md
CHANGED
|
@@ -29,3 +29,18 @@ configs:
|
|
| 29 |
- split: train
|
| 30 |
path: data/train-*
|
| 31 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 29 |
- split: train
|
| 30 |
path: data/train-*
|
| 31 |
---
|
| 32 |
+
|
| 33 |
+
Downloaded from SIGNOR 4.0 on 2/4/2026: https://signor.uniroma2.it/downloads.php
|
| 34 |
+
|
| 35 |
+
Used the Complete Dataset for Homo sapiens
|
| 36 |
+
|
| 37 |
+
- Filtered to entries where TYPEA=='protein' and TYPEB=='protein'
|
| 38 |
+
- Filtered to entries where DIRECT=='YES'
|
| 39 |
+
- Dropped entries where EFFECT=='uknown'
|
| 40 |
+
- Dropped a single entry where EFFECT=='form complex'
|
| 41 |
+
- Filtered to only include the shown columns (renamed)
|
| 42 |
+
- Dropped duplicates
|
| 43 |
+
|
| 44 |
+
Assigned a label of 2 to entries whose EFFECT was 'binding', as well as entries for which the invert appeared (If IDA=X and IDB=Y, when the inverse exists in the dataset, those get a label of 2)
|
| 45 |
+
|
| 46 |
+
The rest of the entries received a label of 1. The IDs for those entries were then inverted and added to the dataset. The inverted pairs received a label of 0.
|