denovoAssemblage/index.js

"use strict";
/*
 * Require in path : zgrep, grep
 */
var __awaiter = (this && this.__awaiter) || function (thisArg, _arguments, P, generator) {
    function adopt(value) { return value instanceof P ? value : new P(function (resolve) { resolve(value); }); }
    return new (P || (P = Promise))(function (resolve, reject) {
        function fulfilled(value) { try { step(generator.next(value)); } catch (e) { reject(e); } }
        function rejected(value) { try { step(generator["throw"](value)); } catch (e) { reject(e); } }
        function step(result) { result.done ? resolve(result.value) : adopt(result.value).then(fulfilled, rejected); }
        step((generator = generator.apply(thisArg, _arguments || [])).next());
    });
};
var __importDefault = (this && this.__importDefault) || function (mod) {
    return (mod && mod.__esModule) ? mod : { "default": mod };
};
Object.defineProperty(exports, "__esModule", { value: true });
exports.denovoAssemblage = void 0;
const child_process_1 = require("child_process");
const os_1 = __importDefault(require("os"));
const path_1 = __importDefault(require("path"));
const fs_1 = __importDefault(require("fs"));
const async_exec = (prog, args, onData, onErr) => {
    return new Promise((resolve, reject) => {
        const child = (0, child_process_1.spawn)(prog, args, { shell: true });
        child.stdout.on('data', data => onData(data.toString().trim()));
        child.stderr.on('data', data => onErr(data.toString().trim()));
        child.on('error', err => reject(err));
        child.on('exit', code => resolve(code));
    });
};
const rmList = (list) => {
    return Promise.all(list.map(e => fs_1.default.promises.rm(e, { recursive: true })));
};
const isDone = (rsltDir) => {
    let done = false;
    if (fs_1.default.existsSync(path_1.default.join(rsltDir, 'contigs.fasta'))) {
        if (fs_1.default.statSync(path_1.default.join(rsltDir, 'contigs.fasta')).size > 1) {
            done = true;
        }
    }
    return done;
};
const denovoAssemblage = (reads, rnames, spadesPath, log) => {
    return new Promise((resolve, reject) => __awaiter(void 0, void 0, void 0, function* () {
        let readsIn = [];
        let isPairedEnd = false;
        const threads = String(os_1.default.cpus().length);
        if (Array.isArray(reads)) {
            if (reads.length > 2) {
                reject('Only R1 and R2 path are required');
            }
            else {
                console.log('Assuming paired end reads');
                isPairedEnd = true;
                readsIn = reads;
            }
        }
        else {
            readsIn = [reads];
        }
        // Prepare the fastq files for spades inputs
        const rnamesSel = '\'' + rnames.join('\\|') + '\'';
        const tmpSubReads = [];
        let isfq = [];
        for (const R of readsIn) {
            const parsedPath = path_1.default.parse(R);
            const greper = parsedPath.ext === '.gz' ? 'zgrep' : 'grep';
            isfq.push(R.match(/\.fq/) ? true : false);
            const tmp = path_1.default.join(os_1.default.tmpdir(), (+new Date) + '_' + parsedPath.name + (isfq[isfq.length - 1] ? '.fq' : '.fa'));
            tmpSubReads.push(tmp);
            yield async_exec(greper, [rnamesSel, R, '-A3', '--no-group-separator', '>', tmp], log, log);
        }
        const correction = (isfq.filter(e => e).length !== isfq.length || isfq.filter(e => e).length === 0) ? '--only-assembler' : '';
        const args = isPairedEnd ? ['-1', tmpSubReads[0], '-2', tmpSubReads[1]] : ['-s', tmpSubReads[0]];
        const rsltDir = path_1.default.join(os_1.default.tmpdir(), (+new Date) + '_spades');
        yield async_exec(spadesPath, ['-t', threads, '--isolate', correction, ...args, '-o', rsltDir], log, log);
        if (!isDone(rsltDir) && isPairedEnd) {
            yield rmList([rsltDir]);
            console.log('Trying meta');
            yield async_exec(spadesPath, ['-t', threads, '--meta', correction, ...args, '-o', rsltDir], log, log);
            if (!isDone(rsltDir)) {
                yield rmList([rsltDir, ...tmpSubReads]);
                reject('No convergence');
            }
        }
        else if (!isDone(rsltDir) && !isPairedEnd) {
            yield rmList([rsltDir, ...tmpSubReads]);
            reject('No convergence');
        }
        else if (isDone(rsltDir)) {
            yield rmList(tmpSubReads);
            resolve(rsltDir);
        }
    }));
};
exports.denovoAssemblage = denovoAssemblage;
/*
(async()=>{
    const spadesPath = '/home/thomas/NGS/tools/SPAdes-3.15.0-Linux/bin/spades.py'
    const reads2 = ['/home/thomas/Documents/Programmes/ttest/R1r.fq', '/home/thomas/Documents/Programmes/ttest/R2r.fq']
    const rnames = [
        'A00680:166:HYCYGDMXX:1:1356:7663:10723',
        'A00680:166:HYCYGDMXX:1:2475:28284:20384',
        'A00680:166:HYCYGDMXX:1:1274:10646:6339',
        'A00680:166:HYCYGDMXX:1:1314:10303:30921',
        'A00680:166:HYCYGDMXX:1:1365:11966:21277',
        'A00680:166:HYCYGDMXX:1:1425:21151:6778',
        'A00680:166:HYCYGDMXX:1:2288:25301:5822',
        'A00680:166:HYCYGDMXX:1:2312:24758:12587',
        'A00680:166:HYCYGDMXX:1:2410:17110:13808',
        'A00680:166:HYCYGDMXX:1:2410:17761:15468',
        'A00680:166:HYCYGDMXX:1:2425:19777:2237',
        'A00680:166:HYCYGDMXX:1:2429:13575:20149',
        'A00680:166:HYCYGDMXX:2:1374:20971:34773',
        'A00680:166:HYCYGDMXX:2:2350:16920:14293',
        'A00680:166:HYCYGDMXX:1:2274:12707:16736',
        'A00680:166:HYCYGDMXX:1:2274:12717:16720',
        'A00680:166:HYCYGDMXX:2:1329:3839:27367',
        'A00680:166:HYCYGDMXX:2:1330:1416:7827'
    ]
    console.log(await denovoAssemblage(reads2, rnames, spadesPath, console.log));
})()
*/