update script

README.md

@@ -17,6 +17,8 @@ It can be used for fine-tuning a language model.
 
 The data solely uses data from PDBind-cn.
 
+Contacts are calculated at three cut-off distances: 5, 8 and 11A.
+
 ### Use the already preprocessed data
 
 Load a test/train split using

pdbbind.ipynb

@@ -579,7 +579,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "5bfa7c0e-fa50-40ab-ad24-a03a8bf01eb2",
+   "id": "02addbaf-094d-4e40-8904-4fe306647298",
    "metadata": {},
    "outputs": [],
    "source": []

pdbbind.py

@@ -20,7 +20,7 @@ punctuation_regex  = r"""(\(|\)|\.|=|#|-|\+|\\|\/|:|~|@|\?|>>?|\*|\$|\%[0-9]{2}|
 # tokenization regex (Schwaller)
 molecule_regex = r"""(\[[^\]]+]|Br?|Cl?|N|O|S|P|F|I|b|c|n|o|s|p|\(|\)|\.|=|#|-|\+|\\|\/|:|~|@|\?|>>?|\*|\$|\%[0-9]{2}|[0-9])"""
 
-cutoff = 5
+cutoff = int(sys.argv[1])
 max_seq = 2046 # = 2048 - 2 (accounting for [CLS] and [SEP])
 max_smiles = 510 # = 512 - 2
 chunk_size = '1G'

pdbbind.slurm

@@ -6,4 +6,6 @@
 #SBATCH -N 2
 #SBATCH --ntasks-per-node=16
 
-srun python pdbbind.py
+srun python pdbbind.py 5
+srun python pdbbind.py 8
+srun python pdbbind.py 11