molstar
Version:
A comprehensive macromolecular library.
253 lines (252 loc) • 10.3 kB
JavaScript
#!/usr/bin/env node
"use strict";
/**
* Copyright (c) 2018-2020 mol* contributors, licensed under MIT, See LICENSE file for more info.
*
* @author Alexander Rose <alexander.rose@weirdbyte.de>
*/
Object.defineProperty(exports, "__esModule", { value: true });
const tslib_1 = require("tslib");
const argparse = tslib_1.__importStar(require("argparse"));
const util = tslib_1.__importStar(require("util"));
const path = tslib_1.__importStar(require("path"));
const fs = tslib_1.__importStar(require("fs"));
require('util.promisify').shim();
const writeFile = util.promisify(fs.writeFile);
const db_1 = require("../../mol-data/db");
const set_1 = require("../../mol-util/set");
const map_1 = require("../../mol-util/map");
const mmcif_extras_1 = require("../../mol-io/reader/cif/schema/mmcif-extras");
const ccd_extras_1 = require("../../mol-io/reader/cif/schema/ccd-extras");
const util_1 = require("./util");
function ccbKey(compId, atomId1, atomId2) {
return atomId1 < atomId2 ? `${compId}:${atomId1}-${atomId2}` : `${compId}:${atomId2}-${atomId1}`;
}
function ccaKey(compId, atomId) {
return `${compId}:${atomId}`;
}
function addChemCompBondToSet(set, ccb) {
for (let i = 0, il = ccb._rowCount; i < il; ++i) {
set.add(ccbKey(ccb.comp_id.value(i), ccb.atom_id_1.value(i), ccb.atom_id_2.value(i)));
}
return set;
}
function addChemCompAtomToSet(set, cca) {
for (let i = 0, il = cca._rowCount; i < il; ++i) {
set.add(ccaKey(cca.comp_id.value(i), cca.atom_id.value(i)));
}
return set;
}
function checkAddingBondsFromPVCD(pvcd) {
const ccbSetByParent = (0, map_1.DefaultMap)(() => new Set());
for (const k in pvcd) {
const { chem_comp, chem_comp_bond } = pvcd[k];
if (chem_comp_bond._rowCount) {
const parentIds = chem_comp.mon_nstd_parent_comp_id.value(0);
if (parentIds.length === 0) {
const set = ccbSetByParent.getDefault(chem_comp.id.value(0));
addChemCompBondToSet(set, chem_comp_bond);
}
else {
for (let i = 0, il = parentIds.length; i < il; ++i) {
const parentId = parentIds[i];
const set = ccbSetByParent.getDefault(parentId);
addChemCompBondToSet(set, chem_comp_bond);
}
}
}
}
for (const k in pvcd) {
const { chem_comp, chem_comp_atom, chem_comp_bond } = pvcd[k];
if (chem_comp_bond._rowCount) {
const parentIds = chem_comp.mon_nstd_parent_comp_id.value(0);
if (parentIds.length > 0) {
for (let i = 0, il = parentIds.length; i < il; ++i) {
const entryBonds = addChemCompBondToSet(new Set(), chem_comp_bond);
const entryAtoms = addChemCompAtomToSet(new Set(), chem_comp_atom);
const extraBonds = set_1.SetUtils.difference(ccbSetByParent.get(parentIds[i]), entryBonds);
extraBonds.forEach(bk => {
const [a1, a2] = bk.split('|');
if (entryAtoms.has(a1) && entryAtoms.has(a2)) {
console.error(`Adding all PVCD bonds would wrongly add bond ${bk} for ${k}`);
}
});
}
}
}
}
}
function checkAddingAtomsFromPVCD(pvcd) {
const ccaSetByParent = (0, map_1.DefaultMap)(() => new Set());
for (const k in pvcd) {
const { chem_comp, chem_comp_atom } = pvcd[k];
if (chem_comp_atom._rowCount) {
const parentIds = chem_comp.mon_nstd_parent_comp_id.value(0);
if (parentIds.length === 0) {
const set = ccaSetByParent.getDefault(chem_comp.id.value(0));
addChemCompAtomToSet(set, chem_comp_atom);
}
else {
for (let i = 0, il = parentIds.length; i < il; ++i) {
const parentId = parentIds[i];
const set = ccaSetByParent.getDefault(parentId);
addChemCompAtomToSet(set, chem_comp_atom);
}
}
}
}
}
async function createBonds(ccd, pvcd, atomsRequested) {
const ccbSet = new Set();
const comp_id = [];
const atom_id_1 = [];
const atom_id_2 = [];
const value_order = [];
const pdbx_aromatic_flag = [];
const pdbx_stereo_config = [];
const molstar_protonation_variant = [];
function addBonds(compId, ccb, protonationVariant) {
for (let i = 0, il = ccb._rowCount; i < il; ++i) {
const atomId1 = ccb.atom_id_1.value(i);
const atomId2 = ccb.atom_id_2.value(i);
const k = ccbKey(compId, atomId1, atomId2);
if (!ccbSet.has(k)) {
atom_id_1.push(atomId1);
atom_id_2.push(atomId2);
comp_id.push(compId);
value_order.push(ccb.value_order.value(i));
pdbx_aromatic_flag.push(ccb.pdbx_aromatic_flag.value(i));
pdbx_stereo_config.push(ccb.pdbx_stereo_config.value(i));
molstar_protonation_variant.push(protonationVariant ? 'Y' : 'N');
ccbSet.add(k);
}
}
}
// check adding bonds from PVCD
checkAddingBondsFromPVCD(pvcd);
// add bonds from PVCD
for (const k in pvcd) {
const { chem_comp, chem_comp_bond } = pvcd[k];
if (chem_comp_bond._rowCount) {
const parentIds = chem_comp.mon_nstd_parent_comp_id.value(0);
if (parentIds.length === 0) {
addBonds(chem_comp.id.value(0), chem_comp_bond, false);
}
else {
for (let i = 0, il = parentIds.length; i < il; ++i) {
addBonds(parentIds[i], chem_comp_bond, true);
}
}
}
}
// add bonds from CCD
for (const k in ccd) {
const { chem_comp, chem_comp_bond } = ccd[k];
if (chem_comp_bond._rowCount) {
addBonds(chem_comp.id.value(0), chem_comp_bond, false);
}
}
const bondTable = db_1.Table.ofArrays(mmcif_extras_1.mmCIF_chemCompBond_schema, {
comp_id, atom_id_1, atom_id_2, value_order,
pdbx_aromatic_flag, pdbx_stereo_config, molstar_protonation_variant
});
const bondDatabase = db_1.Database.ofTables(CCB_TABLE_NAME, { chem_comp_bond: mmcif_extras_1.mmCIF_chemCompBond_schema }, { chem_comp_bond: bondTable });
return { bonds: bondDatabase, atoms: atomsRequested ? createAtoms(ccd, pvcd) : void 0 };
}
function createAtoms(ccd, pvcd) {
const ccaSet = new Set();
const comp_id = [];
const atom_id = [];
const charge = [];
const pdbx_stereo_config = [];
function addAtoms(compId, cca) {
for (let i = 0, il = cca._rowCount; i < il; ++i) {
const atomId = cca.atom_id.value(i);
const k = ccaKey(compId, atomId);
if (!ccaSet.has(k)) {
atom_id.push(atomId);
comp_id.push(compId);
charge.push(cca.charge.value(i));
pdbx_stereo_config.push(cca.pdbx_stereo_config.value(i));
ccaSet.add(k);
}
}
}
// check adding atoms from PVCD
checkAddingAtomsFromPVCD(pvcd);
// add atoms from PVCD
for (const k in pvcd) {
const { chem_comp, chem_comp_atom } = pvcd[k];
if (chem_comp_atom._rowCount) {
const parentIds = chem_comp.mon_nstd_parent_comp_id.value(0);
if (parentIds.length === 0) {
addAtoms(chem_comp.id.value(0), chem_comp_atom);
}
else {
for (let i = 0, il = parentIds.length; i < il; ++i) {
addAtoms(parentIds[i], chem_comp_atom);
}
}
}
}
// add atoms from CCD
for (const k in ccd) {
const { chem_comp, chem_comp_atom } = ccd[k];
if (chem_comp_atom._rowCount) {
addAtoms(chem_comp.id.value(0), chem_comp_atom);
}
}
const atomTable = db_1.Table.ofArrays(ccd_extras_1.ccd_chemCompAtom_schema, {
comp_id, atom_id, charge, pdbx_stereo_config
});
return db_1.Database.ofTables(CCA_TABLE_NAME, { chem_comp_atom: ccd_extras_1.ccd_chemCompAtom_schema }, { chem_comp_atom: atomTable });
}
async function run(out, binary = false, options = util_1.DefaultDataOptions, ccaOut) {
await (0, util_1.ensureDataAvailable)(options);
const ccd = await (0, util_1.readCCD)();
const pvcd = await (0, util_1.readPVCD)();
const { bonds, atoms } = await createBonds(ccd, pvcd, !!ccaOut);
const ccbCif = (0, util_1.getEncodedCif)(CCB_TABLE_NAME, bonds, binary);
if (!fs.existsSync(path.dirname(out))) {
fs.mkdirSync(path.dirname(out));
}
writeFile(out, ccbCif);
if (!!ccaOut) {
const ccaCif = (0, util_1.getEncodedCif)(CCA_TABLE_NAME, atoms, binary);
if (!fs.existsSync(path.dirname(ccaOut))) {
fs.mkdirSync(path.dirname(ccaOut));
}
writeFile(ccaOut, ccaCif);
}
}
const CCB_TABLE_NAME = 'CHEM_COMP_BONDS';
const CCA_TABLE_NAME = 'CHEM_COMP_ATOMS';
const parser = new argparse.ArgumentParser({
add_help: true,
description: 'Create a cif file with one big table of all chem_comp_bond entries from the CCD and PVCD.'
});
parser.add_argument('out', {
help: 'Generated file output path.'
});
parser.add_argument('--forceDownload', '-f', {
action: 'store_true',
help: 'Force download of CCD and PVCD.'
});
parser.add_argument('--binary', '-b', {
action: 'store_true',
help: 'Output as BinaryCIF.'
});
parser.add_argument('--ccaOut', '-a', {
help: 'Optional generated file output path for chem_comp_atom data.',
required: false
});
parser.add_argument('--ccdUrl', '-c', {
help: 'Fetch the CCD from a custom URL. This forces download of the CCD.',
required: false
});
parser.add_argument('--pvcdUrl', '-p', {
help: 'Fetch the PVCD from a custom URL. This forces download of the PVCD.',
required: false
});
const args = parser.parse_args();
run(args.out, args.binary, { forceDownload: args.forceDownload, ccdUrl: args.ccdUrl, pvcdUrl: args.pvcdUrl }, args.ccaOut);