feat(octochemdb): add searchInsilicoSpectra by masses and mf

packages/ms-spectrum/src/MSComparator.js

@@ -145,14 +145,17 @@ function normalizeAndCacheData(cache, dataXY, options = {}) {
 }
 
 function returnSimilarity(aligned, options) {
-  if (options.minNbCommonPeaks) {
-    let commonPeaks = 0;
-    for (let i = 0; i < aligned.ys[0].length; i++) {
-      if (aligned.ys[0][i] !== 0 && aligned.ys[1][i] !== 0) {
-        commonPeaks++;
-      }
+  let commonPeaks = 0;
+  for (let i = 0; i < aligned.ys[0].length; i++) {
+    if (aligned.ys[0][i] !== 0 && aligned.ys[1][i] !== 0) {
+      commonPeaks++;
     }
-    if (commonPeaks < options.minNbCommonPeaks) return 0;
+  }
+  if (
+    commonPeaks === 0 ||
+    (options.minNbCommonPeaks && commonPeaks < options.minNbCommonPeaks)
+  ) {
+    return 0;
   }
 
   const vector1 = new Float64Array(aligned.x.length);

packages/octochemdb/src/OctoChemDB.js

@@ -5,7 +5,8 @@ import { compoundsFromMF } from './compoundsFromMF.js';
 import { massSpectra } from './massSpectra.js';
 import { mfsFromEMs } from './mfsFromEMs.js';
 import { pubmedCompounds } from './pubmedCompounds.js';
-import { searchInSilicoSpectra } from './searchInSilicoSpectra.js';
+import { searchInSilicoSpectraByMF } from './searchInSilicoSpectraByMF.js';
+import { searchInSilicoSpectraByMasses } from './searchInSilicoSpectraByMasses.js';
 import { searchMasses } from './searchMasses.js';
 
 export class OctoChemDB {
@@ -45,8 +46,18 @@ export class OctoChemDB {
     return massSpectra({ baseURL: this.baseURL, ...options });
   }
 
-  async searchInSilicoSpectra(options = {}) {
-    return searchInSilicoSpectra({ baseURL: this.baseURL, ...options });
+  async searchInSilicoSpectraByMF(spectrum, mf, options = {}) {
+    return searchInSilicoSpectraByMF(spectrum, mf, {
+      baseURL: this.baseURL,
+      ...options,
+    });
+  }
+
+  async searchInSilicoSpectraByMasses(spectrum, masses, options = {}) {
+    return searchInSilicoSpectraByMasses(spectrum, masses, {
+      baseURL: this.baseURL,
+      ...options,
+    });
   }
 
   async pubmedCompounds(pubmedID, options = {}) {

packages/octochemdb/src/__tests__/searchInSilicoSpectra.test.js → packages/octochemdb/src/__tests__/searchInSilicoSpectraByMF.test.js

@@ -1,14 +1,14 @@
 import { readFileSync } from 'fs';
 import { join } from 'path';
 
-import { searchInSilicoSpectra } from '../searchInSilicoSpectra.js';
+import { searchInSilicoSpectraByMF } from '../searchInSilicoSpectraByMF.js';
 
 test('searchInsilicoSpectra', async () => {
   const mdma = JSON.parse(
     readFileSync(join(__dirname, 'data/mdma.json'), 'utf8'),
   );
 
-  const results = await searchInSilicoSpectra(mdma, 'C11H15NO2', {
+  const results = await searchInSilicoSpectraByMF(mdma, 'C11H15NO2', {
     precision: 50,
   });
   expect(results.length).toBeGreaterThan(5);

packages/octochemdb/src/__tests__/searchInSilicoSpectraByMasses.test.js

@@ -0,0 +1,25 @@
+import { readFileSync } from 'fs';
+import { join } from 'path';
+
+import { Spectrum } from 'ms-spectrum';
+
+import { OctoChemDB } from '../OctoChemDB.js';
+
+test('searchInSilicoSpectraByMasses', async () => {
+  const mdma = JSON.parse(
+    readFileSync(join(__dirname, 'data/mdma.json'), 'utf8'),
+  );
+  // by default peaks over 1%
+  const spectrum = new Spectrum(mdma).getPeaksAsDataXY();
+  const octoChemDB = new OctoChemDB();
+
+  const results = await octoChemDB.searchInSilicoSpectraByMasses(
+    spectrum,
+    194.1174,
+    {
+      precision: 20,
+      ionizations: 'H+',
+    },
+  );
+  expect(results.length).toBeGreaterThan(5);
+});

packages/octochemdb/src/searchInSilicoSpectra.js → packages/octochemdb/src/searchInSilicoSpectraByMF.js

@@ -3,10 +3,10 @@ import { MSComparator } from 'ms-spectrum';
 import { fetchJSON } from './utils/fetchJSON.js';
 
 /**
- *
+ * @param {object} spectrum - The experimental spectrum to compare to
+ * @param {string} mf - Molecular formula of the non ionized molecule
  * @param {object} [options={}]
  * @param {string} [options.baseURL='https://octochemdb.cheminfo.org/'] - URL of the webservice
- * @param {number|string|number[]} [options.masses] - Observed fragment masses
  * @param {string} [options.mf] - Molecular formula of the non ionized molecule (not available for inSilicoFragments)
  * @param {number} [options.precision=100] - Precision (accuracy) of the monoisotopic mass in ppm
  * @param {number} [options.limit=1000] - Maximal number of entries to return
@@ -16,7 +16,7 @@ import { fetchJSON } from './utils/fetchJSON.js';
  * @param {string} [options.route='inSilicoFragments/v1/search'] - Route to use
  */
 
-export async function searchInSilicoSpectra(spectrum, mf, options = {}) {
+export async function searchInSilicoSpectraByMF(spectrum, mf, options = {}) {
   const {
     route = 'inSilicoFragments/v1/search',
     baseURL = 'https://octochemdb.cheminfo.org/',
@@ -56,8 +56,8 @@ export async function searchInSilicoSpectra(spectrum, mf, options = {}) {
     );
     if (similarity > 0) {
       finals.push({
+        ...result,
         similarity,
-        data: result.data,
       });
     }
   }

packages/octochemdb/src/searchInSilicoSpectraByMasses.js

@@ -0,0 +1,77 @@
+import { MSComparator } from 'ms-spectrum';
+
+import { searchWithIonizations } from './utils/searchWithIonizations.js';
+
+/**
+ * @param {object} spectrum - The experimental spectrum to compare to
+ * @param {number|string|number[]} [masses] - List of experimental monoisotopic mass
+ * @param {object} [options={}]
+ * @param {string} [options.baseURL='https://octochemdb.cheminfo.org/'] - URL of the webservice
+ * @param {string} [options.mf] - Molecular formula of the non ionized molecule (not available for inSilicoFragments)
+ * @param {number} [options.precision=100] - Precision (accuracy) of the monoisotopic mass in ppm
+ * @param {string} [options.ionizations="H+"] - Comma separated list of allowed ionizations
+ * @param {number} [options.limit=1000] - Maximal number of entries to return
+ * @param {number} [options.massPower=1000] - Maximal number of entries to return
+ * @param {number} [options.massPower=3] - High power will give more weight to the mass. If you would prefer to observe fragments you should use a number less than 1
+ * @param {number} [options.intensityPower=0.6] - How important is the intensity. By default we don't give to much importance to it
+ * @param {string} [options.route='inSilicoFragments/v1/search'] - Route to use
+ */
+
+export async function searchInSilicoSpectraByMasses(
+  spectrum,
+  masses,
+  options = {},
+) {
+  const {
+    route = 'inSilicoFragments/v1/search',
+    baseURL = 'https://octochemdb.cheminfo.org/',
+    precision = 100,
+    limit = 10000,
+    massPower = 3,
+    intensityPower = 0.6,
+    ionizations = 'H+',
+  } = options;
+
+  if (!route) {
+    throw new Error('route is mandatory');
+  }
+  if (!masses) {
+    throw new Error('masses is mandatory');
+  }
+
+  const realURL = new URL(route, baseURL).toString();
+
+  let results = await searchWithIonizations({
+    precision,
+    realURL,
+    masses,
+    limit,
+    ionizations,
+  });
+
+  const msComparator = new MSComparator({
+    delta: (mass) => mass * 1e-6 * precision,
+    massPower,
+    intensityPower,
+  });
+
+  const finals = [];
+
+  // filter MDMA
+  // results = results.filter(result => result.data.ocl.idCode === 'dg~D@MBdie]v\\kahHBjh@@')
+
+  for (const result of results) {
+    const similarity = msComparator.getSimilarityToMasses(
+      spectrum,
+      result.data.spectrum.data.x,
+    );
+    if (similarity >= 0) {
+      finals.push({
+        ...result,
+        similarity,
+      });
+    }
+  }
+  finals.sort((a, b) => b.similarity - a.similarity);
+  return finals;
+}

packages/octochemdb/src/searchMasses.js

@@ -12,24 +12,24 @@ import { parseMasses } from './utils/parseMasses.js';
  * @param {number} [options.precision=100] - Precision (accuracy) of the monoisotopic mass in ppm
  * @param {number} [options.limit=1000] - Maximal number of entries to return
  * @param {string} [options.modifications=''] - Comma separated list of allowed modifications
- * @param {string} [options.url='massBank/v1/search'] - Route to use
+ * @param {string} [options.route='massBank/v1/search'] - Route to use
  * @param {string} [options.link='https://massbank.eu/MassBank/RecordDisplay?id='] - Link to the database source
  */
 
 export async function searchMasses(options = {}) {
   const {
-    url = '',
+    route = 'massBank/v1/search',
     baseURL = 'https://octochemdb.cheminfo.org/',
     precision = 100,
     limit = 1000,
     mf = '',
     link = '',
   } = options;
 
-  if (url === '') {
-    throw new Error('url is mandatory');
+  if (route === '') {
+    throw new Error('route is mandatory');
   }
-  const realURL = new URL(url, baseURL).toString();
+  const realURL = new URL(route, baseURL).toString();
   const masses = parseMasses(options.masses);
 
   const modifications = preprocessIonizations(options.modifications);

packages/octochemdb/src/utils/searchWithIonizations.js

@@ -12,7 +12,7 @@ export async function searchWithIonizations(options) {
     fields,
     precision = 100,
     limit = 1000,
-    searchParams,
+    searchParams = {},
     ranges,
   } = options;
 
@@ -30,7 +30,6 @@ export async function searchWithIonizations(options) {
     precision,
     ionizations,
   });
-
   const promises = [];
   for (let ionization of ionizations) {
     for (let mass of masses) {