From 2dd17e715ac16b5ba7c0679721a65d3b3b9514bc Mon Sep 17 00:00:00 2001
From: dominiquesydow <dominique.sydow@posteo.de>
Date: Sun, 17 Apr 2022 16:08:39 +0100
Subject: [PATCH] Update kissim scripts to run residue subsets

---
 scripts/kissim_subsets.sh                     | 28 +++++++++++++++++++
 ...and_subset.py => kissim_subsets_ligand.py} | 18 ++++++------
 2 files changed, 38 insertions(+), 8 deletions(-)
 create mode 100644 scripts/kissim_subsets.sh
 rename scripts/{kissim_ligand_subset.py => kissim_subsets_ligand.py} (86%)

diff --git a/scripts/kissim_subsets.sh b/scripts/kissim_subsets.sh
new file mode 100644
index 0000000..5d40621
--- /dev/null
+++ b/scripts/kissim_subsets.sh
@@ -0,0 +1,28 @@
+# @dominiquesydow
+# Subset and compare fingerprints
+# - DFG-in focused residues
+# - Martin et al. residues
+# 
+# Run this script only after having results from running the main script
+# https://github.com/volkamerlab/kissim_app/blob/master/scripts/kissim_encode_compare.sh
+#
+# Usage
+# -----
+# cd /path/to/kissim_app
+# bash kissim_subsets.sh
+
+KISSIM_APP="."
+NCORES=7
+
+# DFG-in focused residues
+for name_subset in dfg_in martin_set
+do
+    RESULTS=$KISSIM_APP/results/dfg_in_subset_${name_subset}
+    echo $RESULTS
+    echo "Select fingerprint subset..."
+    kissim subset -i $RESULTS/fingerprints_normalized.json -o $RESULTS/fingerprints_subset.json -s $name_subset
+    echo "Done."
+    echo "Compare fingerprints..."
+    kissim compare -i $RESULTS/fingerprints_subset.json -o $RESULTS -c $NCORES
+    echo "Done."
+done
\ No newline at end of file
diff --git a/scripts/kissim_ligand_subset.py b/scripts/kissim_subsets_ligand.py
similarity index 86%
rename from scripts/kissim_ligand_subset.py
rename to scripts/kissim_subsets_ligand.py
index 7bdc56a..16f3d09 100644
--- a/scripts/kissim_ligand_subset.py
+++ b/scripts/kissim_subsets_ligand.py
@@ -1,5 +1,5 @@
 """
-Generate KiSSim data for 
+Generate KiSSim data for ligand-based residue subsets.
 """
 
 from pathlib import Path
@@ -10,9 +10,10 @@
 
 PATH_RESULTS_DFG_IN = PATH_RESULTS / "dfg_in"
 
+
 def main(ligand_expo_id, path_fp_in=PATH_RESULTS_DFG_IN / "fingerprints_normalized.json"):
     """
-    Generate KiSSim fingerprint distances for a subset version of the fingerprint that 
+    Generate KiSSim fingerprint distances for a subset version of the fingerprint that
     is based only on residues that show interactions with co-crystallized structures
     (use KLIFS IFPS for this).
 
@@ -36,22 +37,23 @@ def main(ligand_expo_id, path_fp_in=PATH_RESULTS_DFG_IN / "fingerprints_normaliz
     # Subset fingerprints
     print("Subset fingerprints...")
     fingerprint_generator = subset(
-        fingerprints_path = path_fp_in,
+        fingerprints_path=path_fp_in,
         klifs_pocket_residue_subset_type=ligand_expo_id,
-        fingerprints_subset_path = path_out / "fingerprints_normalized.json",
-        klifs_pocket_residue_subset={ligand_expo_id: interacting_residues}
+        fingerprints_subset_path=path_out / "fingerprints_normalized.json",
+        klifs_pocket_residue_subset={ligand_expo_id: interacting_residues},
     )
-    
+
     print("All-against-all comparison...")
     compare(
         fingerprint_generator=fingerprint_generator,
         output_path=path_out,
         feature_weights=None,
-        n_cores=7
+        n_cores=7,
     )
 
+
 if __name__ == "__main__":
     main("IRE")
     main("STI")
     main("DB8")
-    main("B96")
\ No newline at end of file
+    main("B96")