Merge pull request #957 from papillot/fix-pdbqt

Fix PDBQT parser for wrong H elements interpolation

src/parser/pdb-parser.ts

@@ -12,7 +12,7 @@ import StructureParser from './structure-parser'
 import Entity, { EntityTypeString } from '../structure/entity'
 import Unitcell, { UnitcellParams } from '../symmetry/unitcell'
 import Assembly, { AssemblyPart } from '../symmetry/assembly'
-import { WaterNames } from '../structure/structure-constants'
+import { PDBQTSpecialElements, WaterNames } from '../structure/structure-constants'
 import {
   assignSecondaryStructure, buildUnitcellAssembly,
   calculateBonds, calculateChainnames, calculateSecondaryStructure
@@ -301,7 +301,9 @@ class PdbParser extends StructureParser {
 
             if (!isLegacy) {
               if (isPdbqt) {
-                element = line.substr(12, 2).trim()
+                element = line.substr(76, 3).trim()
+                // @ts-expect-error TS limitation on narrowing indexes types with `in`
+                if (element in PDBQTSpecialElements) element = PDBQTSpecialElements[element]
               } else {
                 element = line.substr(76, 2).trim()
                 if (!chainname) {

src/structure/structure-constants.ts

@@ -1105,3 +1105,24 @@ ResidueTypeAtoms[ CgDnaBackboneType ] = {
 }
 
 ResidueTypeAtoms[ UnknownBackboneType ] = {}
+
+// Mappings taken from Meeko: https://github.com/forlilab/Meeko/blob/develop/meeko/utils/autodock4_atom_types_elements.py
+export const PDBQTSpecialElements = {
+  'HD':  'H',
+  'HS':  'H',
+  'A':   'C',
+  'NA':  'N',
+  'NS':  'N',
+  'OA':  'O',
+  'OS':  'O',
+  'SA':  'S',
+  'G0':  'C',
+  'G1':  'C',
+  'G2':  'C',
+  'G3':  'C',
+  'CG0': 'C',
+  'CG1': 'C',
+  'CG2': 'C',
+  'CG3': 'C',
+  'W':   'O'
+}

test/data/1xdn_gln.pdbqt

@@ -0,0 +1,15 @@
+ATOM      1  N   GLN    52      42.237  16.800  35.823  1.00  0.00    -0.066 N 
+ATOM      2  CA  GLN    52      41.667  17.015  34.477  1.00  0.00     0.275 C 
+ATOM      3  C   GLN    52      40.148  17.010  34.446  1.00  0.00     0.249 C 
+ATOM      4  O   GLN    52      39.564  17.006  33.382  1.00  0.00    -0.271 OA
+ATOM      5  CB  GLN    52      42.228  15.967  33.522  1.00  0.00     0.052 C 
+ATOM      6  CG  GLN    52      43.746  16.016  33.446  1.00  0.00     0.105 C 
+ATOM      7  CD  GLN    52      44.292  17.263  32.785  1.00  0.00     0.215 C 
+ATOM      8  OE1 GLN    52      43.587  17.906  32.000  1.00  0.00    -0.274 OA
+ATOM      9  NE2 GLN    52      45.513  17.598  33.072  1.00  0.00    -0.370 N 
+ATOM     10  H2  GLN    52      41.772  15.848  36.146  1.00  0.00     0.275 HD
+ATOM     11  H3  GLN    52      43.141  16.691  35.848  1.00  0.00     0.275 HD
+ATOM     12  H   GLN    52      41.923  17.526  36.434  1.00  0.00     0.275 HD
+ATOM     13 2HE2 GLN    52      46.044  17.022  33.735  1.00  0.00     0.159 HD
+ATOM     14 1HE2 GLN    52      45.958  18.423  32.653  1.00  0.00     0.159 HD
+TER      15      GLN    52 

test/parser/tests-pdbqt-parser.spec.ts

@@ -0,0 +1,26 @@
+import StringStreamer from '../../src/streamer/string-streamer'
+import PdbqtParser from '../../src/parser/pdbqt-parser'
+import { Structure } from '../../src/ngl'
+
+
+import { join } from 'path'
+import * as fs from 'fs'
+
+
+describe('parser/pdb-parser', function () {
+  describe('parsing', function () {
+    it('translates PDBQT special atom types to elements', function() {
+      const file = join(__dirname, '/../data/1xdn_gln.pdbqt')
+      const str = fs.readFileSync(file, 'utf-8')
+      const streamer = new StringStreamer(str)
+      const pdbqtParser = new PdbqtParser(streamer)
+      return pdbqtParser.parse().then(function (structure: Structure) {
+        const ap = structure.getAtomProxy()
+        ap.index = 3
+        expect(ap.element).toBe('O')
+        ap.index = 12
+        expect(ap.element).toBe('H')
+      })
+    })
+  })
+})