@sjcrh/proteinpaint-shared 2.78.0-0

package/package.json

@@ -0,0 +1,18 @@
+{
+  "name": "@sjcrh/proteinpaint-shared",
+  "version": "2.78.0-0",
+  "description": "ProteinPaint code that is shared between server and client-side workspaces",
+  "type": "module",
+  "main": "src/index.js",
+  "exports": {
+    "./*": "./src/*"
+  },
+  "directories": {
+    "test": "test"
+  },
+  "scripts": {
+    "test": "echo \"Error: no test specified\" && exit 1"
+  },
+  "author": "",
+  "license": "ISC"
+}

package/src/bulk.cnv.js

@@ -0,0 +1,86 @@
+////////////////////////////////////
+//
+//  shared between client and server
+//
+////////////////////////////////////
+
+import * as common from './common.js'
+import * as bulk from './bulk.js'
+
+export function parseheader(line, flag) {
+	const header = line.toLowerCase().split('\t')
+	if (header.length <= 1) return 'invalid file header for CNV'
+	const htry = (...lst) => {
+		for (const i of lst) {
+			const j = header.indexOf(i)
+			if (j != -1) return j
+		}
+		return -1
+	}
+	let i = htry('gene')
+	if (i == -1) return 'gene missing from header'
+	header[i] = 'gene'
+	i = htry('cnv')
+	if (i == -1) return 'CNV missing from header'
+	header[i] = 'cnv'
+	i = htry('sample', 'sample_name', 'tumor_sample_barcode')
+	if (i != -1) header[i] = 'sample'
+	i = htry('patient', 'donor', 'target_case_id')
+	if (i != -1) header[i] = 'patient'
+	i = htry('disease')
+	if (i != -1) header[i] = 'disease'
+	i = htry('origin')
+	if (i != -1) header[i] = 'origin'
+	i = htry('sampletype', 'sample type', 'sample_type')
+	if (i != -1) header[i] = 'sampletype'
+	flag.cnv.header = header
+	flag.cnv.loaded = true
+	return false
+}
+
+export function parseline(i, line, flag) {
+	if (line == '' || line[0] == '#') return
+	const lst = line.split('\t')
+	const m = {}
+	for (let j = 0; j < flag.cnv.header.length; j++) {
+		m[flag.cnv.header[j]] = lst[j]
+	}
+	if (!m.gene) {
+		flag.cnv.badlines.push([i, 'missing gene', lst])
+		return
+	}
+	if (!m.cnv) {
+		flag.cnv.badlines.push([i, 'missing cnv value', lst])
+		return
+	}
+	const value = m.cnv.toLowerCase()
+	switch (value) {
+		case 'amplification':
+		case 'gain':
+			m.class = common.mclasscnvgain
+			break
+		case 'deletion':
+		case 'loss':
+			m.class = common.mclasscnvloss
+			break
+		case 'loh':
+			m.class = common.mclasscnvloh
+			break
+		default:
+			flag.cnv.badlines.push([i, 'invalid cnv value: ' + m.cnv, lst])
+			m.class = null
+	}
+	if (!m.class) {
+		return
+	}
+	if (bulk.parsesample(m, flag, i, lst, flag.cnv.badlines)) {
+		return
+	}
+	m.dt = common.dtcnv
+	flag.good++
+	const n = flag.geneToUpper ? m.gene.toUpperCase() : m.gene
+	if (!(n in flag.data)) {
+		flag.data[n] = []
+	}
+	flag.data[n].push(m)
+}

package/src/bulk.del.js

@@ -0,0 +1,124 @@
+////////////////////////////////////
+//
+//  shared between client and server
+//
+////////////////////////////////////
+
+import * as common from './common.js'
+import * as bulk from './bulk.js'
+
+export function parseheader(line, flag) {
+	const header = line.toLowerCase().split('\t')
+	if (header.length <= 1) return 'invalid header line for intragenic deletion'
+	const htry = (...lst) => {
+		for (const i of lst) {
+			const j = header.indexOf(i)
+			if (j != -1) return j
+		}
+		return -1
+	}
+	let i = htry('gene')
+	if (i == -1) return 'gene missing from header'
+	header[i] = 'gene'
+	i = htry(
+		'annovar_isoform',
+		'mrna_accession',
+		'mrna accession',
+		'refseq_mrna_id',
+		'annovar_sj_filter_isoform',
+		'refseq',
+		'isoform'
+	)
+	if (i == -1) return 'isoform missing from header'
+	header[i] = 'isoform'
+	i = htry('rnaposition')
+	if (i != -1) {
+		header[i] = 'rnaposition'
+		i = htry('rnadellength')
+		if (i == -1) return 'rnadellength is required when rnaPosition is used'
+		header[i] = 'rnadellength'
+	}
+	i = htry('chromosome', 'chr')
+	if (i != -1) {
+		header[i] = 'chr'
+		i = htry('chr_start')
+		if (i == -1) return 'chr_start is required when chr is used'
+		header[i] = 'chrpos1'
+		i = htry('chr_stop')
+		if (i == -1) return 'chr_stop is required when chr is used'
+		header[i] = 'chrpos2'
+	}
+
+	i = htry('sample', 'sample_name', 'tumor_sample_barcode')
+	if (i != -1) header[i] = 'sample'
+	i = htry('patient', 'donor', 'target_case_id')
+	if (i != -1) header[i] = 'patient'
+	i = htry('disease')
+	if (i != -1) header[i] = 'disease'
+	i = htry('origin')
+	if (i != -1) header[i] = 'origin'
+	i = htry('sampletype', 'sample type', 'sample_type')
+	if (i != -1) header[i] = 'sampletype'
+	flag.del.header = header
+	flag.del.loaded = true
+	return false
+}
+
+export function parseline(i, line, flag) {
+	if (line == '' || line[0] == '#') return
+	const lst = line.split('\t')
+	const m = {}
+	for (let j = 0; j < flag.del.header.length; j++) {
+		if (lst[j] == undefined) break
+		m[flag.del.header[j]] = lst[j]
+	}
+	if (!m.gene) {
+		flag.del.badlines.push([i, 'missing gene', lst])
+		return
+	}
+	if (m.rnaposition) {
+		let v = Number.parseInt(m.rnaposition)
+		if (Number.isNaN(v) || v < 0) {
+			flag.del.badlines.push([i, 'invalid rnaPosition value', lst])
+			return
+		}
+		m.rnaposition = v
+		if (!m.rnadellength) {
+			flag.del.badlines.push([i, 'missing rnaDellength value', lst])
+			return
+		}
+		v = Number.parseInt(m.rnadellength)
+		if (Number.isNaN(v) || v < 0) {
+			flag.del.badlines.push([i, 'invalid rnaDellength value', lst])
+			return
+		}
+		m.rnadellength = v
+	}
+	if (m.chr) {
+		let v = Number.parseInt(m.chrpos1)
+		if (Number.isNaN(v) || v < 0) {
+			flag.del.badlines.push([i, 'invalid chr_start value', lst])
+			return
+		}
+		m.chrpos1 = v
+		v = Number.parseInt(m.chrpos2)
+		if (Number.isNaN(v) || v < 0) {
+			flag.del.badlines.push([i, 'invalid chr_stop value', lst])
+			return
+		}
+		m.chrpos2 = v
+	}
+
+	if (bulk.parsesample(m, flag, i, lst, flag.del.badlines)) {
+		return
+	}
+	m.dt = common.dtdel
+	m.class = common.mclassdel
+	m.mname = 'DEL'
+	flag.good++
+	var n = flag.geneToUpper ? m.gene.toUpperCase() : m.gene
+	if (!(n in flag.data)) {
+		flag.data[n] = []
+	}
+	flag.data[n].push(m)
+}

package/src/bulk.itd.js

@@ -0,0 +1,123 @@
+////////////////////////////////////
+//
+//  shared between client and server
+//
+////////////////////////////////////
+
+import * as common from './common.js'
+import * as bulk from './bulk.js'
+
+export function parseheader(line, flag) {
+	const header = line.toLowerCase().split('\t')
+	if (header.length <= 1) return 'invalid header line for ITD'
+	const htry = (...lst) => {
+		for (const i of lst) {
+			const j = header.indexOf(i)
+			if (j != -1) return j
+		}
+		return -1
+	}
+	let i = htry('gene')
+	if (i == -1) return 'gene missing from header'
+	header[i] = 'gene'
+	i = htry(
+		'annovar_isoform',
+		'mrna_accession',
+		'mrna accession',
+		'refseq_mrna_id',
+		'annovar_sj_filter_isoform',
+		'refseq',
+		'isoform'
+	)
+	if (i == -1) return 'isoform missing from header'
+	header[i] = 'isoform'
+	i = htry('rnaposition')
+	if (i != -1) {
+		header[i] = 'rnaposition'
+		i = htry('rnaduplength')
+		if (i == -1) return 'rnaduplength is required when rnaposition is present'
+		header[i] = 'rnaduplength'
+	}
+	i = htry('chromosome', 'chr')
+	if (i != -1) {
+		header[i] = 'chr'
+		i = htry('chr_start')
+		if (i == -1) return 'chr_start is required when chr is present'
+		header[i] = 'chrpos1'
+		i = htry('chr_stop')
+		if (i == -1) return 'chr_stop is required when chr is present'
+		header[i] = 'chrpos2'
+	}
+
+	i = htry('sample', 'sample_name', 'tumor_sample_barcode')
+	if (i != -1) header[i] = 'sample'
+	i = htry('patient', 'donor', 'target_case_id')
+	if (i != -1) header[i] = 'patient'
+	i = htry('disease')
+	if (i != -1) header[i] = 'disease'
+	i = htry('origin')
+	if (i != -1) header[i] = 'origin'
+	i = htry('sampletype', 'sample type', 'sample_type')
+	if (i != -1) header[i] = 'sampletype'
+	flag.itd.header = header
+	flag.itd.loaded = true
+	return false
+}
+
+export function parseline(i, line, flag) {
+	if (line == '' || line[0] == '#') return
+	const lst = line.split('\t')
+	const m = {}
+	for (let j = 0; j < flag.itd.header.length; j++) {
+		if (lst[j] == undefined) break
+		m[flag.itd.header[j]] = lst[j]
+	}
+	if (!m.gene) {
+		flag.itd.badlines.push([i, 'missing gene', lst])
+		return
+	}
+	if (m.rnaposition) {
+		let v = Number.parseInt(m.rnaposition)
+		if (Number.isNaN(v) || v < 0) {
+			flag.itd.badlines.push([i, 'invalid rnaPosition value', lst])
+			return
+		}
+		m.rnaposition = v
+		if (!m.rnaduplength) {
+			flag.itd.badlines.push([i, 'missing rnaDuplength value', lst])
+			return
+		}
+		v = Number.parseInt(m.rnaduplength)
+		if (Number.isNaN(v) || v < 0) {
+			flag.itd.badlines.push([i, 'invalid rnaDuplength value', lst])
+			return
+		}
+		m.rnaduplength = v
+	}
+	if (m.chr) {
+		let v = Number.parseInt(m.chrpos1)
+		if (Number.isNaN(v) || v < 0) {
+			flag.itd.badlines.push([i, 'invalid chr_start value', lst])
+			return
+		}
+		m.chrpos1 = v
+		v = Number.parseInt(m.chrpos2)
+		if (Number.isNaN(v) || v < 0) {
+			flag.itd.badlines.push([i, 'invalid chr_stop value', lst])
+			return
+		}
+		m.chrpos2 = v
+	}
+	if (bulk.parsesample(m, flag, i, lst, flag.itd.badlines)) {
+		return
+	}
+	m.dt = common.dtitd
+	m.class = common.mclassitd
+	m.mname = 'ITD'
+	flag.good++
+	var n = flag.geneToUpper ? m.gene.toUpperCase() : m.gene
+	if (!(n in flag.data)) {
+		flag.data[n] = []
+	}
+	flag.data[n].push(m)
+}

package/src/bulk.js

@@ -0,0 +1,197 @@
+////////////////////////////////////
+//
+//  shared between client and server
+//
+////////////////////////////////////
+
+import * as common from './common.js'
+
+export default {}
+
+export function init_bulk_flag(genome) {
+	if (!genome) {
+		return null
+	}
+	const mclasslabel2key = {}
+	for (const n in common.mclass) {
+		mclasslabel2key[common.mclass[n].label.toUpperCase()] = n
+	}
+	return {
+		genome: genome,
+		mclasslabel2key: mclasslabel2key,
+		data: {},
+		sample2disease: {}, // (proof) k: sample, v: disease
+		// will only record this when origin is used
+		patient2st: {},
+		// k: patient, v: { k: sampletype, v: sample }
+		// new sample names always override old
+		good: 0,
+		geneToUpper: true, // option to not force uppercase on gene names
+		snv: {
+			loaded: false,
+			header: null,
+			badlines: [],
+			// jinghui: based on missense/silent ratio of entire dataset to decide whether to include silent when importing...
+			// hard-coded class codes
+			missense: 0,
+			silent: 0
+		},
+		svjson: {
+			loaded: false,
+			header: null,
+			badlines: []
+		},
+		fusion: {
+			loaded: false,
+			header: null,
+			badlines: [],
+			original: []
+		},
+		sv: {
+			loaded: false,
+			header: null,
+			badlines: [],
+			original: []
+		},
+		cnv: {
+			loaded: false,
+			header: null,
+			badlines: []
+		},
+		itd: {
+			loaded: false,
+			header: null,
+			badlines: []
+		},
+		del: {
+			loaded: false,
+			header: null,
+			badlines: []
+		},
+		truncation: {
+			loaded: false,
+			header: null,
+			badlines: []
+		}
+	}
+}
+
+export function parsesample(m, flag, i, lst, badline) {
+	let variantorigin = common.moriginsomatic
+	if (m.sampletype) {
+		const s = m.sampletype.toLowerCase()
+		switch (s) {
+			case 'relapse':
+				variantorigin = common.moriginrelapse
+				break
+			case 'germline':
+				variantorigin = common.morigingermline
+				break
+			case 'somatic':
+			case 'diagnosis':
+				break
+		}
+		if (m.sample) {
+			if (m.patient) {
+				// good
+			} else {
+				m.patient = m.sample + ' ' + m.sampletype
+			}
+		} else {
+			if (m.patient) {
+				m.sample = m.patient + ' ' + m.sampletype
+			} else {
+				// neither sample or patient, will quit later
+			}
+		}
+	} else {
+		if (m.patient) {
+			if (m.sample) {
+				m.sampletype = m.sample
+			} else {
+				m.sample = m.sampletype = m.patient
+			}
+		} else {
+			if (m.sample) {
+				m.sampletype = m.sample
+			} else {
+				// no patient/sample, will quit later
+			}
+		}
+	}
+	if (m.origin) {
+		// override existing variantorigin
+		const s = m.origin.toLowerCase()
+		switch (s) {
+			case 'r':
+			case 'relapse':
+				variantorigin = common.moriginrelapse
+				m.isrim2 = true
+				break
+			case 'g':
+			case 'germline':
+				variantorigin = common.morigingermline
+				m.isrim1 = true
+				break
+			case 'gp':
+			case 'germline pathogenic':
+				variantorigin = common.morigingermlinepathogenic
+				m.isrim1 = true
+				break
+			case 'gnp':
+			case 'germline nonpathogenic':
+			case 'germline non-pathogenic':
+				variantorigin = common.morigingermlinenonpathogenic
+				m.isrim1 = true
+				break
+			case 's':
+			case 'somatic':
+			case 'diagnosis':
+				variantorigin = common.moriginsomatic
+				break
+		}
+	}
+	m.origin = variantorigin
+
+	if (!m.sample && !m.patient) {
+		// will not go into sample table
+		return
+	}
+
+	const nopatientname = 'no patient/individual name'
+	let p
+	if (m.patient) {
+		if (!flag.patient2st[m.patient]) {
+			flag.patient2st[m.patient] = {}
+		}
+		flag.patient2st[m.patient][m.sampletype] = m.sample
+	} else {
+		if (!flag.patient2st[nopatientname]) {
+			flag.patient2st[nopatientname] = {}
+		}
+		flag.patient2st[nopatientname][m.sampletype] = m.sample
+	}
+
+	if (m.sample) {
+		if (m.disease) {
+			if (m.sample in flag.sample2disease) {
+				if (m.disease != flag.sample2disease[m.sample]) {
+					flag.snv.badlines.push([
+						i,
+						'conflict of disease types for sample "' +
+							m.sample +
+							'": ' +
+							m.disease +
+							', ' +
+							flag.sample2disease[m.sample],
+						lst
+					])
+					return true
+				}
+			} else {
+				flag.sample2disease[m.sample] = m.disease
+			}
+		}
+	}
+	return false
+}