ml-gsd - npm Package Compare versions

Comparing version 5.0.2 to 6.0.0

lib/index.js

@@ -6,3 +6,3 @@ 'use strict';
 var SG = require('ml-savitzky-golay-generalized');
-var mlOptimizeLorentzian = require('ml-optimize-lorentzian');
+var mlSpectraFitting = require('ml-spectra-fitting');
 
@@ -15,5 +15,6 @@ function _interopDefaultLegacy (e) { return e && typeof e === 'object' && 'default' in e ? e : { 'default': e }; }
  * Global spectra deconvolution
- * @param {Array<number>} x - Independent variable
- * @param {Array<number>} yIn - Dependent variable
- * @param {object} [options] - Options object
+ * @param {object} data - Object data with x and y arrays
+ * @param {Array<number>} [data.x] - Independent variable
+ * @param {Array<number>} [data.y] - Dependent variable
+ * @param {object} [options={}] - Options object
  * @param {object} [options.sgOptions] - Options object for Savitzky-Golay filter. See https://github.com/mljs/savitzky-golay-generalized#options
@@ -34,3 +35,3 @@ * @param {number} [options.sgOptions.windowSize = 9] - points to use in the approximations
  */
-function gsd(x, yIn, options = {}) {
+function gsd(data, options = {}) {
   let {
@@ -51,2 +52,4 @@ noiseLevel,
 
+  let { y: yIn, x } = data;
+
   const y = yIn.slice();
@@ -355,6 +358,20 @@ let equalSpaced = isEqualSpaced(x);
 
-function optimizePeaks(peakList, x, y, options = {}) {
+/**
+ * Optimize the position (x), max intensity (y), full width at half maximum (width)
+ * and the ratio of gaussian contribution (mu) if it's required. It supports three kind of shapes: gaussian, lorentzian and pseudovoigt
+ * @param {object} data - An object containing the x and y data to be fitted.
+ * @param {Array} peakList - A list of initial parameters to be optimized. e.g. coming from a peak picking [{x, y, width}].
+ * @param {object} [options = {}] -
+ * @param {string} [options.kind = 'gaussian'] - kind of shape used to fitting, lorentzian, gaussian and pseudovoigt are supported.
+ * @param {number} [options.factorWidth = 4] - times of width to group peaks.
+ * @param {object} [options.joinPeaks = true] - if true the peaks could be grouped if the separation between them are inside of a range of factorWidth * width
+ * @param {object} [options.optimizationOptions] - options of ml-levenberg-marquardt optimization package.
+ */
+const kindSupported = ['gaussian', 'lorentzian', 'pseudovoigt'];
+
+function optimizePeaks(data, peakList, options = {}) {
   const {
     functionName = 'gaussian',
     factorWidth = 4,
+    joinPeaks = true,
     optimizationOptions = {
@@ -367,9 +384,10 @@ damping: 1.5,
 
+  let { x, y } = data;
+
+  checkFuncName(functionName, optimizationOptions);
+
   let lastIndex = [0];
-  let groups = groupPeaks(peakList, factorWidth);
+  let groups = groupPeaks(peakList, factorWidth, joinPeaks);
+
   let result = [];
-  let factor = 1;
-  if (functionName === 'gaussian') {
-    factor = 1.17741;
-  } // From https://en.wikipedia.org/wiki/Gaussian_function#Properties
   let sampling;
@@ -387,24 +405,10 @@ for (let i = 0; i < groups.length; i++) {
       );
-      if (sampling[0].length > 5) {
-        let optPeaks = [];
-        if (functionName === 'gaussian') {
-          optPeaks = mlOptimizeLorentzian.optimizeGaussianSum(sampling, peaks, optimizationOptions);
-        } else {
-          if (functionName === 'lorentzian') {
-            optPeaks = mlOptimizeLorentzian.optimizeLorentzianSum(
-              sampling,
-              peaks,
-              optimizationOptions,
-            );
-          }
-        }
-
+      if (sampling.x.length > 5) {
+        let { peaks: optPeaks } = mlSpectraFitting.optimize(sampling, peaks, {
+          kind: functionName,
+          lmOptions: optimizationOptions,
+        });
         for (let j = 0; j < optPeaks.length; j++) {
-          let { parameters } = optPeaks[j];
-          result.push({
-            x: parameters[0],
-            y: parameters[1],
-            width: parameters[2] * factor,
-            index: peaks[j].index,
-          });
+          optPeaks[j].index = peaks.index;
+          result.push(optPeaks[j]);
         }
@@ -423,27 +427,10 @@ }
 
-      if (sampling[0].length > 5) {
-        let fitResult = [];
-        if (functionName === 'gaussian') {
-          fitResult = mlOptimizeLorentzian.optimizeSingleGaussian(
-            [sampling[0], sampling[1]],
-            peaks,
-            optimizationOptions,
-          );
-        } else {
-          if (functionName === 'lorentzian') {
-            fitResult = mlOptimizeLorentzian.optimizeSingleLorentzian(
-              [sampling[0], sampling[1]],
-              peaks,
-              optimizationOptions,
-            );
-          }
-        }
-
-        let { parameters } = fitResult;
-        result.push({
-          x: parameters[0],
-          y: parameters[1],
-          width: parameters[2] * factor,
-          index: peaks.index,
-        }); // From https://en.wikipedia.org/wiki/Gaussian_function#Properties}
+      if (sampling.x.length > 5) {
+        let fitResult = mlSpectraFitting.optimize(sampling, [peaks], {
+          kind: functionName,
+          lmOptions: optimizationOptions,
+        });
+        let { peaks: optPeaks } = fitResult;
+        optPeaks[0].index = peaks.index;
+        result.push(optPeaks[0]);
       }
@@ -485,6 +472,6 @@ }
   lastIndex[0] = index;
-  return [sampleX, sampleY];
+  return { x: sampleX, y: sampleY };
 }
 
-function groupPeaks(peakList, nL) {
+function groupPeaks(peakList, nL, joinPeaks) {
   let group = [];
@@ -498,4 +485,4 @@ let groups = [];
     if (
-      Math.abs(peakList[i].x - limits[0]) <
-      nL * peakList[i].width + limits[1]
+      joinPeaks &&
+      Math.abs(peakList[i].x - limits[0]) < nL * peakList[i].width + limits[1]
     ) {
@@ -519,3 +506,2 @@ // Add the peak to the group
       groups.push({ limits: limits, group: group });
-      // var optmimalPeak = fitSpectrum(group,limits,spectrum);
       group = [peakList[i]];
@@ -556,6 +542,21 @@ limits = [peakList[i].x, nL * peakList[i].width];
 
+function checkFuncName(functionName, optimizationOptions) {
+  let kind = functionName.toLowerCase().replace(/[^a-z]/g, '');
+  let isSupported = kindSupported.some((ks) => ks === kind);
+  if (isSupported) {
+    optimizationOptions.kind = kind;
+  } else {
+    throw new Error(
+      `Kind of function unsupported. Just these kind are supported: ${kindSupported.join(
+        ', ',
+      )}`,
+    );
+  }
+}
+
 /**
  * This function try to join the peaks that seems to belong to a broad signal in a single broad peak.
- * @param peakList
- * @param options
+ * @param {Array} peakList - A list of initial parameters to be optimized. e.g. coming from a peak picking [{x, y, width}].
+ * @param {object} [options = {}] -
+ * @param {number} options.width - width limit to join peaks.
  */
@@ -579,9 +580,8 @@ function joinBroadPeaks(peakList, options = {}) {
 
-  let candidates = [[broadLines[0].x, broadLines[0].y]];
-  let indexes = [broadLines[0].index];
-
+  let candidates = { x: [broadLines[0].x], y: [broadLines[0].y] };
+  let indexes = [0];
   for (let i = 1; i < broadLines.length; i++) {
-    // console.log(broadLines[i-1].x+" "+broadLines[i].x);
     if (Math.abs(broadLines[i - 1].x - broadLines[i].x) < width) {
-      candidates.push([broadLines[i].x, broadLines[i].y]);
+      candidates.x.push(broadLines[i].x);
+      candidates.y.push(broadLines[i].y);
       if (broadLines[i].y > max) {
@@ -591,23 +591,25 @@ max = broadLines[i].y;
       }
-      indexes.push(broadLines[i].index);
+      indexes.push(i);
       count++;
     } else {
       if (count > 2) {
-        let fitted = mlOptimizeLorentzian.optimizeSingleLorentzian(candidates, {
-          x: broadLines[maxI].x,
-          y: max,
-          width: Math.abs(
-            candidates[0][0] - candidates[candidates.length - 1][0],
-          ),
-        });
-        let { parameters } = fitted;
-        peakList.push({
-          x: parameters[0],
-          y: parameters[1],
-          width: parameters[2],
-          index: Math.floor(
-            indexes.reduce((a, b) => a + b, 0) / indexes.length,
-          ),
-          soft: false,
-        });
+        let fitted = mlSpectraFitting.optimize(
+          candidates,
+          [
+            {
+              x: broadLines[maxI].x,
+              y: max,
+              width: Math.abs(
+                candidates.x[0] - candidates.x[candidates.x.length - 1],
+              ),
+            },
+          ],
+          { kind: 'lorentzian' },
+        );
+        let { peaks: peak } = fitted;
+        peak[0].index = Math.floor(
+          indexes.reduce((a, b) => a + b, 0) / indexes.length,
+        );
+        peak[0].soft = false;
+        peakList.push(peak[0]);
       } else {
@@ -619,3 +621,3 @@ // Put back the candidates to the signals list
       }
-      candidates = [[broadLines[i].x, broadLines[i].y]];
+      candidates = { x: [broadLines[i].x], y: [broadLines[i].y] };
       indexes = [i];
@@ -627,3 +629,2 @@ max = broadLines[i].y;
   }
-
   peakList.sort(function (a, b) {
@@ -630,0 +631,0 @@ return a.x - b.x;

package.json

 {
   "name": "ml-gsd",
-  "version": "5.0.2",
+  "version": "6.0.0",
   "description": "Global Spectra Deconvolution",
@@ -15,3 +15,3 @@ "main": "lib/index.js",
   "scripts": {
-    "build": "cheminfo-build --entry src/index.js --root GSD",
+    "build": "rollup -c && cheminfo-build --entry src/index.js --root GSD",
     "eslint": "eslint src --cache",
@@ -44,26 +44,34 @@ "eslint-fix": "npm run eslint -- --fix",
   "homepage": "https://github.com/mljs/global-spectral-deconvolution",
+  "jest": {
+    "testEnvironment": "node"
+  },
+  "prettier": {
+    "arrowParens": "always",
+    "semi": true,
+    "singleQuote": true,
+    "tabWidth": 2,
+    "trailingComma": "all"
+  },
   "devDependencies": {
-    "@babel/plugin-transform-modules-commonjs": "^7.10.4",
+    "@babel/plugin-transform-modules-commonjs": "^7.12.1",
     "chemcalc": "^3.4.1",
-    "cheminfo-tools": "^1.23.3",
-    "eslint": "^7.9.0",
-    "eslint-config-cheminfo": "^3.0.0",
-    "eslint-plugin-import": "^2.22.0",
-    "eslint-plugin-jest": "^23.20.0",
+    "cheminfo-build": "^1.1.8",
+    "eslint": "^7.12.1",
+    "eslint-config-cheminfo": "^5.2.2",
+    "eslint-plugin-import": "^2.22.1",
+    "eslint-plugin-jest": "^24.1.0",
     "eslint-plugin-prettier": "^3.1.4",
-    "jest": "^26.4.2",
+    "esm": "^3.2.25",
+    "jest": "^26.6.3",
+    "mf-global": "^1.3.0",
     "ml-stat": "^1.3.3",
     "prettier": "^2.1.2",
-    "rollup": "^2.28.2",
-    "spectrum-generator": "^4.0.2",
+    "rollup": "^2.33.1",
+    "spectrum-generator": "^4.4.0",
     "xy-parser": "^3.0.0"
   },
   "dependencies": {
-    "ml-optimize-lorentzian": "^0.2.0",
-    "ml-savitzky-golay-generalized": "2.0.2"
-  },
-  "jest": {
-    "verbose": true,
-    "testURL": "http://localhost/"
+    "ml-savitzky-golay-generalized": "2.0.2",
+    "ml-spectra-fitting": "^0.5.0"
   }
 }

README.md

 # global-spectral-deconvolution
 
 [![NPM version][npm-image]][npm-url]
-[![build status][travis-image]][travis-url]
 [![David deps][david-image]][david-url]
@@ -10,2 +9,6 @@ [![npm download][download-image]][download-url]
 
+`gsd`is using an algorithm that is searching for inflection points to determine the position of peaks and the width of the peaks are between the 2 inflection points. The result of GSD yield to an array of object containing {x, y and width}. However this width is based on the inflection point and may be different from the 'fwhm' (Full Width Half Maximum).
+
+The second algorithm (`optimize`) will optimize the width as a FWHM to match the original peak. After optimization the width with therefore be always FWHM whichever is the function used. 
+
 ## [API documentation](http://mljs.github.io/global-spectral-deconvolution/)
@@ -53,3 +56,3 @@
 
-### GSD.post.broadenPeaks(peakList, {factor=2, overlap=false})
+### GSD.broadenPeaks(peakList, {factor=2, overlap=false})
 
@@ -59,5 +62,5 @@ We enlarge the peaks and add the properties from and to.
 
-### GSD.post.joinBroadPeaks
+### GSD.joinBroadPeaks
 
-### GSD.post.optimizePeaks
+### GSD.optimizePeaks
 
@@ -67,31 +70,24 @@ ## Example
 ```js
-var CC = require('chemcalc');
-var Stat = require('ml-stat');
-var peakPicking = require('ml-gsd');
+import { IsotopicDistribution } from 'mf-global';
+import { gsd, optimizePeaks } from '../src';
 
-var spectrum = CC.analyseMF('Cl2.Br2', {
-  isotopomers: 'arrayXXYY',
-  fwhm: 0.01,
-  gaussianWidth: 11
+// generate a sample spectrum of the form {x:[], y:[]}
+const data = new IsotopicDistribution('C').getGaussian();
+
+let peaks = gsd(data, {
+  noiseLevel: 0,
+  minMaxRatio: 0.00025, // Threshold to determine if a given peak should be considered as a noise
+  realTopDetection: true,
+  maxCriteria: true, // inverted:false
+  smoothY: false,
+  sgOptions: { windowSize: 7, polynomial: 3 },
 });
-var xy = spectrum.arrayXXYY;
-var x = xy[0];
-var y = xy[1];
-//Just a fake noiseLevel
-var noiseLevel =
-  Stat.array.median(
-    y.filter(function(a) {
-      return a > 0;
-    })
-  ) * 3;
+console.log(peaks); // array of peaks {x,y,width}, width = distance between inflection points
+// GSD
 
-var options = {
-  noiseLevel: noiseLevel,
-  minMaxRatio: 0,
-  broadRatio: 0,
-  smoothY: false,
-  realTopDetection: true
-};
-var result = peakPicking.gsd(x, y, options);
-result = peakPicking.post.optimizePeaks(result, x, y, 1, 'gaussian');
+let optimized = optimizePeaks(data, peaks);
+console.log(optimized); // array of peaks {x,y,width}, width = FWHM
+
+
+
 ```
@@ -105,4 +101,2 @@
 [npm-url]: https://npmjs.org/package/ml-gsd
-[travis-image]: https://img.shields.io/travis/mljs/global-spectral-deconvolution/master.svg?style=flat-square
-[travis-url]: https://travis-ci.org/mljs/global-spectral-deconvolution
 [david-image]: https://img.shields.io/david/mljs/global-spectral-deconvolution.svg?style=flat-square
@@ -109,0 +103,0 @@ [david-url]: https://david-dm.org/mljs/global-spectral-deconvolution

src/__tests__/bigLowS2NSpectrum.js

@@ -16,8 +16,11 @@ /**
 
-    let pp = gsd(spectrum[0], spectrum[1], {
-      noiseLevel: 57000.21889405926, // 1049200.537996172/2,
-      minMaxRatio: 0.01,
-      broadRatio: 0.0025,
-      sgOptions: { windowSize: 13, polynomial: 3 },
-    });
+    let pp = gsd(
+      { x: spectrum[0], y: spectrum[1] },
+      {
+        noiseLevel: 57000.21889405926, // 1049200.537996172/2,
+        minMaxRatio: 0.01,
+        broadRatio: 0.0025,
+        sgOptions: { windowSize: 13, polynomial: 3 },
+      },
+    );
 
@@ -24,0 +27,0 @@ joinBroadPeaks(pp, { width: 0.25 });

src/__tests__/broadNMR.js

@@ -12,11 +12,14 @@ import { readFileSync } from 'fs';
     );
-    let result = gsd(spectrum[0], spectrum[1], {
-      noiseLevel: 1049200.537996172 / 2,
-      minMaxRatio: 0.01,
-      broadRatio: 0.0025,
-      sgOptions: {
-        windowSize: 9,
-        polynomial: 3,
+    let result = gsd(
+      { x: spectrum[0], y: spectrum[1] },
+      {
+        noiseLevel: 1049200.537996172 / 2,
+        minMaxRatio: 0.01,
+        broadRatio: 0.0025,
+        sgOptions: {
+          windowSize: 9,
+          polynomial: 3,
+        },
       },
-    });
+    );
     joinBroadPeaks(result, { width: 0.25 });
@@ -23,0 +26,0 @@ expect(result).toHaveLength(14);

src/__tests__/ethylvinylether.js

@@ -11,13 +11,14 @@ import { readFileSync } from 'fs';
     );
-    let result = gsd(spectrum[0], spectrum[1], {
-      // noiseLevel: 1049200.537996172,
-      minMaxRatio: 0.03,
-      smoothY: false,
-      realTopDetection: true,
-      sgOptions: { windowSize: 5, polynomial: 3 },
-    });
-    // console.log(spectrum[1][13223]);
-    // console.log(result);
+    let result = gsd(
+      { x: spectrum[0], y: spectrum[1] },
+      {
+        // noiseLevel: 1049200.537996172,
+        minMaxRatio: 0.03,
+        smoothY: false,
+        realTopDetection: true,
+        sgOptions: { windowSize: 5, polynomial: 3 },
+      },
+    );
     expect(result).toHaveLength(21);
   });
 });

src/__tests__/infraRed.js

@@ -11,3 +11,3 @@ import { readFileSync } from 'fs';
     );
-    gsd(spectrum.x, spectrum.y, {
+    gsd(spectrum, {
       noiseLevel: 32,
@@ -14,0 +14,0 @@ minMaxRatio: 0.03,

src/__tests__/massPeakPicking.js

@@ -28,10 +28,13 @@ import { gsd, optimizePeaks } from '..';
   it('Check result', () => {
-    let result = gsd(x, y, {
-      noiseLevel: noiseLevel,
-      minMaxRatio: 0,
-      broadRatio: 0,
-      smoothY: false,
-      realTopDetection: true,
-    });
-    result = optimizePeaks(result, x, y, {
+    let result = gsd(
+      { x, y },
+      {
+        noiseLevel: noiseLevel,
+        minMaxRatio: 0,
+        broadRatio: 0,
+        smoothY: false,
+        realTopDetection: true,
+      },
+    );
+    result = optimizePeaks({ x, y }, result, {
       factorWidth: 1,
@@ -38,0 +41,0 @@ functionName: 'gaussian',

src/__tests__/simple.js

 import { gsd } from '..';
 
 describe('Simple test cases', () => {
-  let X = [];
-  let Y = [];
+  let x = [];
+  let y = [];
   for (let i = 0; i < 10; i++) {
-    X.push(X.length);
-    Y.push(0);
+    x.push(x.length);
+    y.push(0);
   }
   for (let i = 0; i <= 10; i++) {
-    X.push(X.length);
-    Y.push(i > 5 ? 10 - i : i);
+    x.push(x.length);
+    y.push(i > 5 ? 10 - i : i);
   }
   for (let i = 0; i < 10; i++) {
-    X.push(X.length);
-    Y.push(0);
+    x.push(x.length);
+    y.push(0);
   }
 
   it('gsd not realtop', () => {
-    let peaks = gsd(X, Y, {
-      realTopDetection: false,
-      smoothY: true,
-      sgOptions: {
-        windowSize: 5,
-        polynomial: 3,
+    let peaks = gsd(
+      { x, y },
+      {
+        realTopDetection: false,
+        smoothY: true,
+        sgOptions: {
+          windowSize: 5,
+          polynomial: 3,
+        },
       },
-    });
+    );
 
@@ -35,12 +38,15 @@ expect(peaks[0].y).toBeCloseTo(4.657, 3);
   it('gsd not realtop asymetric', () => {
-    let Y2 = Y.slice(0);
+    let Y2 = y.slice(0);
     Y2[14] = 5;
-    let peaks = gsd(X, Y2, {
-      realTopDetection: false,
-      smoothY: true,
-      sgOptions: {
-        windowSize: 5,
-        polynomial: 3,
+    let peaks = gsd(
+      { x, y: Y2 },
+      {
+        realTopDetection: false,
+        smoothY: true,
+        sgOptions: {
+          windowSize: 5,
+          polynomial: 3,
+        },
       },
-    });
+    );
 
@@ -68,12 +74,15 @@ expect(peaks).toStrictEqual([
   it('gsd realtop', () => {
-    let Y2 = Y.slice();
+    let Y2 = y.slice();
     Y2[14] = 5;
-    let peaks = gsd(X, Y2, {
-      realTopDetection: true,
-      smoothY: false,
-      sgOptions: {
-        windowSize: 5,
-        polynomial: 3,
+    let peaks = gsd(
+      { x, y: Y2 },
+      {
+        realTopDetection: true,
+        smoothY: false,
+        sgOptions: {
+          windowSize: 5,
+          polynomial: 3,
+        },
       },
-    });
+    );
     expect(peaks).toStrictEqual([
@@ -80,0 +89,0 @@ {

src/__tests__/simulated.js

@@ -1,4 +0,4 @@
-import { gsd } from '..';
+import { gsd, optimizePeaks } from '..';
 
-const { SpectrumGenerator } = require('spectrum-generator');
+const { generateSpectrum } = require('spectrum-generator');
 
@@ -8,11 +8,10 @@ describe('Global spectra deconvolution with simulated spectra', () => {
   it('Should provide the right result ...', () => {
-    const sg = new SpectrumGenerator({ start: 0, end: 100, pointsPerUnit: 10 });
+    const peaks = [
+      { x: 0.1, y: 0.2, width: 0.1 },
+      { x: -0.1, y: 0.2, width: 0.3 },
+    ];
 
-    sg.addPeak([20, 100], { width: 5 });
-    sg.addPeak([50, 50], { width: 5 });
-    sg.addPeak([70, 20], { width: 5 });
+    const data = generateSpectrum(peaks, { from: -1, to: 1, nbPoints: 101 });
 
-    let spectrum = sg.getSpectrum();
-
-    let result = gsd(spectrum.x, spectrum.y, {
+    let peakList = gsd(data, {
       minMaxRatio: 0,
@@ -23,6 +22,11 @@ realTopDetection: false,
 
-    expect(result[0]).toMatchObject({ x: 20, y: 100 });
-    expect(result[1]).toMatchObject({ x: 50, y: 50 });
-    expect(result[2]).toMatchObject({ x: 70, y: 20 });
+    let optPeaks = optimizePeaks(data, peakList, {});
+
+    expect(optPeaks[0].x).toBeCloseTo(-0.1, 2);
+    expect(optPeaks[0].y).toBeCloseTo(0.2, 2);
+    expect(optPeaks[0].width).toBeCloseTo(0.3, 2);
+    expect(optPeaks[1].x).toBeCloseTo(0.1, 2);
+    expect(optPeaks[1].y).toBeCloseTo(0.2, 2);
+    expect(optPeaks[1].width).toBeCloseTo(0.1, 2);
   });
 });

src/__tests__/test.js

@@ -18,8 +18,11 @@ import { readFileSync } from 'fs';
     );
-    let result = gsd(spectrum[0], spectrum[1], {
-      // noiseLevel: 0.001,
-      minMaxRatio: 0,
-      realTopDetection: true,
-      smoothY: false,
-    });
+    let result = gsd(
+      { x: spectrum[0], y: spectrum[1] },
+      {
+        // noiseLevel: 0.001,
+        minMaxRatio: 0,
+        realTopDetection: true,
+        smoothY: false,
+      },
+    );
 
@@ -53,8 +56,11 @@ expect(result[0].x).toBeCloseTo(24, 2);
     }
-    let ans = gsd(times, tic, {
-      noiseLevel: 0,
-      realTopDetection: false,
-      smoothY: false,
-      sgOptions: { windowSize: 5, polynomial: 3 },
-    });
+    let ans = gsd(
+      { x: times, y: tic },
+      {
+        noiseLevel: 0,
+        realTopDetection: false,
+        smoothY: false,
+        sgOptions: { windowSize: 5, polynomial: 3 },
+      },
+    );
 
@@ -61,0 +67,0 @@ expect(ans).toHaveLength(10);

src/__tests__/ubiquitin.js

@@ -21,3 +21,3 @@ import { readFileSync } from 'fs';
 
-    let result = gsd(spectrum.x, spectrum.y, {
+    let result = gsd(spectrum, {
       noiseLevel: noiseLevel,
@@ -24,0 +24,0 @@ minMaxRatio: 0.0,

src/__tests__/zero.js

 import { gsd } from '..';
 
 test('Simple test cases', () => {
-  let X = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15];
-  let Y = [0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0];
-  let peaks = gsd(X, Y, {
-    noiseLevel: 0,
-    minMaxRatio: 0.00025, // Threshold to determine if a given peak should be considered as a noise
-    realTopDetection: true,
-    maxCriteria: true, // inverted:false
-    smoothY: false,
-    sgOptions: { windowSize: 7, polynomial: 3 },
-  });
+  let x = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15];
+  let y = [0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0];
+  let peaks = gsd(
+    { x, y },
+    {
+      noiseLevel: 0,
+      minMaxRatio: 0.00025, // Threshold to determine if a given peak should be considered as a noise
+      realTopDetection: true,
+      maxCriteria: true, // inverted:false
+      smoothY: false,
+      sgOptions: { windowSize: 7, polynomial: 3 },
+    },
+  );
   expect(peaks[0].x).toBe(8);
 });

src/gsd.js

@@ -5,5 +5,6 @@ import SG from 'ml-savitzky-golay-generalized';
  * Global spectra deconvolution
- * @param {Array<number>} x - Independent variable
- * @param {Array<number>} yIn - Dependent variable
- * @param {object} [options] - Options object
+ * @param {object} data - Object data with x and y arrays
+ * @param {Array<number>} [data.x] - Independent variable
+ * @param {Array<number>} [data.y] - Dependent variable
+ * @param {object} [options={}] - Options object
  * @param {object} [options.sgOptions] - Options object for Savitzky-Golay filter. See https://github.com/mljs/savitzky-golay-generalized#options
@@ -24,3 +25,3 @@ * @param {number} [options.sgOptions.windowSize = 9] - points to use in the approximations
  */
-export function gsd(x, yIn, options = {}) {
+export function gsd(data, options = {}) {
   let {
@@ -41,2 +42,4 @@ noiseLevel,
 
+  let { y: yIn, x } = data;
+
   const y = yIn.slice();
@@ -43,0 +46,0 @@ let equalSpaced = isEqualSpaced(x);

src/post/joinBroadPeaks.js

@@ -1,7 +0,8 @@
-import { optimizeSingleLorentzian } from 'ml-optimize-lorentzian';
+import { optimize } from 'ml-spectra-fitting';
 
 /**
  * This function try to join the peaks that seems to belong to a broad signal in a single broad peak.
- * @param peakList
- * @param options
+ * @param {Array} peakList - A list of initial parameters to be optimized. e.g. coming from a peak picking [{x, y, width}].
+ * @param {object} [options = {}] -
+ * @param {number} options.width - width limit to join peaks.
  */
@@ -25,9 +26,8 @@ export function joinBroadPeaks(peakList, options = {}) {
 
-  let candidates = [[broadLines[0].x, broadLines[0].y]];
-  let indexes = [broadLines[0].index];
-
+  let candidates = { x: [broadLines[0].x], y: [broadLines[0].y] };
+  let indexes = [0];
   for (let i = 1; i < broadLines.length; i++) {
-    // console.log(broadLines[i-1].x+" "+broadLines[i].x);
     if (Math.abs(broadLines[i - 1].x - broadLines[i].x) < width) {
-      candidates.push([broadLines[i].x, broadLines[i].y]);
+      candidates.x.push(broadLines[i].x);
+      candidates.y.push(broadLines[i].y);
       if (broadLines[i].y > max) {
@@ -37,23 +37,25 @@ max = broadLines[i].y;
       }
-      indexes.push(broadLines[i].index);
+      indexes.push(i);
       count++;
     } else {
       if (count > 2) {
-        let fitted = optimizeSingleLorentzian(candidates, {
-          x: broadLines[maxI].x,
-          y: max,
-          width: Math.abs(
-            candidates[0][0] - candidates[candidates.length - 1][0],
-          ),
-        });
-        let { parameters } = fitted;
-        peakList.push({
-          x: parameters[0],
-          y: parameters[1],
-          width: parameters[2],
-          index: Math.floor(
-            indexes.reduce((a, b) => a + b, 0) / indexes.length,
-          ),
-          soft: false,
-        });
+        let fitted = optimize(
+          candidates,
+          [
+            {
+              x: broadLines[maxI].x,
+              y: max,
+              width: Math.abs(
+                candidates.x[0] - candidates.x[candidates.x.length - 1],
+              ),
+            },
+          ],
+          { kind: 'lorentzian' },
+        );
+        let { peaks: peak } = fitted;
+        peak[0].index = Math.floor(
+          indexes.reduce((a, b) => a + b, 0) / indexes.length,
+        );
+        peak[0].soft = false;
+        peakList.push(peak[0]);
       } else {
@@ -65,3 +67,3 @@ // Put back the candidates to the signals list
       }
-      candidates = [[broadLines[i].x, broadLines[i].y]];
+      candidates = { x: [broadLines[i].x], y: [broadLines[i].y] };
       indexes = [i];
@@ -73,3 +75,2 @@ max = broadLines[i].y;
   }
-
   peakList.sort(function (a, b) {
@@ -76,0 +77,0 @@ return a.x - b.x;

src/post/optimizePeaks.js

@@ -1,12 +0,21 @@
-import {
-  optimizeGaussianSum,
-  optimizeLorentzianSum,
-  optimizeSingleGaussian,
-  optimizeSingleLorentzian,
-} from 'ml-optimize-lorentzian';
+import { optimize } from 'ml-spectra-fitting';
 
-export function optimizePeaks(peakList, x, y, options = {}) {
+/**
+ * Optimize the position (x), max intensity (y), full width at half maximum (width)
+ * and the ratio of gaussian contribution (mu) if it's required. It supports three kind of shapes: gaussian, lorentzian and pseudovoigt
+ * @param {object} data - An object containing the x and y data to be fitted.
+ * @param {Array} peakList - A list of initial parameters to be optimized. e.g. coming from a peak picking [{x, y, width}].
+ * @param {object} [options = {}] -
+ * @param {string} [options.kind = 'gaussian'] - kind of shape used to fitting, lorentzian, gaussian and pseudovoigt are supported.
+ * @param {number} [options.factorWidth = 4] - times of width to group peaks.
+ * @param {object} [options.joinPeaks = true] - if true the peaks could be grouped if the separation between them are inside of a range of factorWidth * width
+ * @param {object} [options.optimizationOptions] - options of ml-levenberg-marquardt optimization package.
+ */
+const kindSupported = ['gaussian', 'lorentzian', 'pseudovoigt'];
+
+export function optimizePeaks(data, peakList, options = {}) {
   const {
     functionName = 'gaussian',
     factorWidth = 4,
+    joinPeaks = true,
     optimizationOptions = {
@@ -19,9 +28,10 @@ damping: 1.5,
 
+  let { x, y } = data;
+
+  checkFuncName(functionName, optimizationOptions);
+
   let lastIndex = [0];
-  let groups = groupPeaks(peakList, factorWidth);
+  let groups = groupPeaks(peakList, factorWidth, joinPeaks);
+
   let result = [];
-  let factor = 1;
-  if (functionName === 'gaussian') {
-    factor = 1.17741;
-  } // From https://en.wikipedia.org/wiki/Gaussian_function#Properties
   let sampling;
@@ -39,24 +49,10 @@ for (let i = 0; i < groups.length; i++) {
       );
-      if (sampling[0].length > 5) {
-        let optPeaks = [];
-        if (functionName === 'gaussian') {
-          optPeaks = optimizeGaussianSum(sampling, peaks, optimizationOptions);
-        } else {
-          if (functionName === 'lorentzian') {
-            optPeaks = optimizeLorentzianSum(
-              sampling,
-              peaks,
-              optimizationOptions,
-            );
-          }
-        }
-
+      if (sampling.x.length > 5) {
+        let { peaks: optPeaks } = optimize(sampling, peaks, {
+          kind: functionName,
+          lmOptions: optimizationOptions,
+        });
         for (let j = 0; j < optPeaks.length; j++) {
-          let { parameters } = optPeaks[j];
-          result.push({
-            x: parameters[0],
-            y: parameters[1],
-            width: parameters[2] * factor,
-            index: peaks[j].index,
-          });
+          optPeaks[j].index = peaks.index;
+          result.push(optPeaks[j]);
         }
@@ -75,27 +71,10 @@ }
 
-      if (sampling[0].length > 5) {
-        let fitResult = [];
-        if (functionName === 'gaussian') {
-          fitResult = optimizeSingleGaussian(
-            [sampling[0], sampling[1]],
-            peaks,
-            optimizationOptions,
-          );
-        } else {
-          if (functionName === 'lorentzian') {
-            fitResult = optimizeSingleLorentzian(
-              [sampling[0], sampling[1]],
-              peaks,
-              optimizationOptions,
-            );
-          }
-        }
-
-        let { parameters } = fitResult;
-        result.push({
-          x: parameters[0],
-          y: parameters[1],
-          width: parameters[2] * factor,
-          index: peaks.index,
-        }); // From https://en.wikipedia.org/wiki/Gaussian_function#Properties}
+      if (sampling.x.length > 5) {
+        let fitResult = optimize(sampling, [peaks], {
+          kind: functionName,
+          lmOptions: optimizationOptions,
+        });
+        let { peaks: optPeaks } = fitResult;
+        optPeaks[0].index = peaks.index;
+        result.push(optPeaks[0]);
       }
@@ -137,6 +116,6 @@ }
   lastIndex[0] = index;
-  return [sampleX, sampleY];
+  return { x: sampleX, y: sampleY };
 }
 
-function groupPeaks(peakList, nL) {
+function groupPeaks(peakList, nL, joinPeaks) {
   let group = [];
@@ -150,4 +129,4 @@ let groups = [];
     if (
-      Math.abs(peakList[i].x - limits[0]) <
-      nL * peakList[i].width + limits[1]
+      joinPeaks &&
+      Math.abs(peakList[i].x - limits[0]) < nL * peakList[i].width + limits[1]
     ) {
@@ -171,3 +150,2 @@ // Add the peak to the group
       groups.push({ limits: limits, group: group });
-      // var optmimalPeak = fitSpectrum(group,limits,spectrum);
       group = [peakList[i]];
@@ -207,1 +185,15 @@ limits = [peakList[i].x, nL * peakList[i].width];
 }
+
+function checkFuncName(functionName, optimizationOptions) {
+  let kind = functionName.toLowerCase().replace(/[^a-z]/g, '');
+  let isSupported = kindSupported.some((ks) => ks === kind);
+  if (isSupported) {
+    optimizationOptions.kind = kind;
+  } else {
+    throw new Error(
+      `Kind of function unsupported. Just these kind are supported: ${kindSupported.join(
+        ', ',
+      )}`,
+    );
+  }
+}

New alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

New author

Supply chain risk

A new npm collaborator published a version of the package for the first time. New collaborators are usually benign additions to a project, but do indicate a change to the security surface area of a package.

Found 1 instance in 1 package

Fixed alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Improved metrics

Total package byte prevSize

5859086

increased by0.05%

Number of package files

29

increased by3.57%

Lines of code

1754

increased by1.45%

Worsened metrics

Dev dependency count

16

increased by14.29%

Number of lines in readme file

101

decreased by-5.61%

Number of medium supply chain risk alerts

1

increased byInfinity%

Dependency changes

• + Addedml-spectra-fitting@^0.5.0

• + Addedd3-random@2.2.2(transitive)

• + Addedesm@3.2.25(transitive)

• + Addedml-peak-shape-generator@0.7.01.0.0(transitive)

• + Addedml-spectra-fitting@0.5.0(transitive)

• + Addedml-xsadd@2.0.0(transitive)

• + Addedobject-hash@2.2.0(transitive)

• + Addedspectrum-generator@4.8.1(transitive)

• - Removedml-optimize-lorentzian@^0.2.0

• - Removedml-optimize-lorentzian@0.2.0(transitive)