@teselagen/bio-parsers 0.4.27 → 0.4.29-beta.1

package/package.json

@@ -1,19 +1,17 @@
 {
   "name": "@teselagen/bio-parsers",
-  "version": "0.4.27",
+  "version": "0.4.29-beta.1",
   "type": "module",
   "dependencies": {
     "@gmod/gff": "^1.2.1",
-    "buffer": "5.7.1",
     "bufferpack": "^0.0.6",
     "color": "3.2.1",
     "fast-xml-parser": "^4.2.5",
     "fflate": "^0.8.0",
     "lodash-es": "^4.17.21",
-    "string_decoder": "^1.3.0",
     "validate.io-nonnegative-integer-array": "^1.0.1",
-    "@teselagen/sequence-utils": "0.3.30",
-    "@teselagen/range-utils": "0.3.13"
+    "@teselagen/sequence-utils": "0.3.32-beta.1",
+    "@teselagen/range-utils": "0.3.14-beta.1"
   },
   "exports": {
     ".": {

package/src/ab1ToJson.js

@@ -121,50 +121,45 @@ const tagDict = {
   colorDataC: { tagName: "DATA", tagNum: 12, typeToReturn: "getShort" }
 };
 
-const correctionAmount = 3;
 // tnr: this function takes in chromData which has 4 traces and a basePos (which describes where in the trace the base call lands)
 // It "normalizes" that data into a baseTraces array so that each base has its own set of that data (having a per-base trace makes insertion/deletion/copy/paste actions all easier)
 function convertBasePosTraceToPerBpTrace(chromData) {
-  const { basePos, aTrace } = chromData;
-  const traceLength = aTrace.length;
-  let startPos = 0;
-  let nextBasePos = basePos[1];
-  let endPos;
-  function setEndPos() {
-    if (nextBasePos) {
-      endPos = startPos + Math.ceil((nextBasePos - startPos) / 2);
-    } else {
-      endPos = traceLength;
-    }
+  const { basePos } = chromData;
+
+  const peakEdges = [0];
+  for (let i = 0; i < basePos.length - 1; i++) {
+    peakEdges.push(Math.ceil((basePos[i] + basePos[i + 1]) / 2));
   }
-  setEndPos();
+  peakEdges.push(chromData.aTrace.length);
+
+  // Trim edges of trace so that the first and last peak traces are roughly symmetric
+  // around the peak
+  const firstBinWidth = peakEdges[1] - peakEdges[0];
+  const secondBinWidth = peakEdges[2] - peakEdges[1];
+  if (firstBinWidth > secondBinWidth) {
+    peakEdges[0] = peakEdges[1] - secondBinWidth;
+  }
+
+  const lastBinWidth =
+    peakEdges[peakEdges.length - 1] - peakEdges[peakEdges.length - 2];
+  const secondLastBinWidth =
+    peakEdges[peakEdges.length - 2] - peakEdges[peakEdges.length - 3];
+  if (lastBinWidth > secondLastBinWidth) {
+    peakEdges[peakEdges.length - 1] =
+      peakEdges[peakEdges.length - 2] + secondLastBinWidth + 1;
+  }
+
   const baseTraces = [];
-  for (let i = 0; i < basePos.length; i++) {
+  for (let i = 0; i < peakEdges.length - 1; i++) {
+    const start = peakEdges[i];
+    const end = peakEdges[i + 1];
     const tracesForType = {
-      aTrace: [],
-      tTrace: [],
-      gTrace: [],
-      cTrace: []
+      aTrace: chromData.aTrace.slice(start, end),
+      tTrace: chromData.tTrace.slice(start, end),
+      gTrace: chromData.gTrace.slice(start, end),
+      cTrace: chromData.cTrace.slice(start, end)
     };
-    baseTraces[i] = tracesForType;
-    [
-      "aTrace",
-      "tTrace",
-      "gTrace",
-      "cTrace"
-      // eslint-disable-next-line no-loop-func
-    ].forEach(type => {
-      const traceForType = tracesForType[type];
-      const traceData = chromData[type];
-      for (let j = startPos; j < endPos + correctionAmount; j++) {
-        traceForType.push(traceData[j] || 0);
-      }
-    });
-    if (i !== basePos.length - 1) {
-      startPos = endPos + correctionAmount;
-      nextBasePos = basePos[i + 2];
-      setEndPos();
-    }
+    baseTraces.push(tracesForType);
   }
 
   return {

package/src/anyToJson.js

@@ -10,7 +10,7 @@ import { tidyUpSequenceData } from "@teselagen/sequence-utils";
 import geneiousXmlToJson from "./geneiousXmlToJson";
 import jbeiXmlToJson from "./jbeiXmlToJson";
 import { unzipSync } from "fflate";
-
+import fastqToJson from "./fastqToJson";
 /**
  * takes in file content string and its file name and determines what parser it needs to be sent to.
  * The file is parsed to our old JSON schema and after it goes through an intermediate step where we convert that json to our new schema
@@ -74,6 +74,9 @@ async function anyToJson(fileContentStringOrFileObj, options) {
   if (/^(fasta|fas|fa|fna|ffn|faa)$/.test(ext)) {
     // FASTA
     return fastaToJson(fileContentString, options);
+  } else if (/^(fastq)$/.test(ext)) {
+    // FASTQ
+    return fastqToJson(fileContentString, options);
   } else if (/^(gb|gbk)$/.test(ext)) {
     // GENBANK
     return genbankToJson(fileContentString, options);
@@ -182,8 +185,8 @@ function getUtf8StringFromFile(file, { emulateBrowser } = {}) {
     return Buffer.isBuffer(file)
       ? file.toString("utf-8")
       : Buffer.isBuffer(file.buffer)
-      ? file.buffer.toString("utf-8")
-      : file;
+        ? file.buffer.toString("utf-8")
+        : file;
   }
   const reader = new window.FileReader();
   reader.readAsText(file, "UTF-8");
@@ -204,8 +207,8 @@ function getUint8ArrayFromFile(file, { emulateBrowser } = {}) {
     return Buffer.isBuffer(file)
       ? new Uint8Array(file)
       : Buffer.isBuffer(file.buffer)
-      ? new Uint8Array(file.buffer)
-      : file;
+        ? new Uint8Array(file.buffer)
+        : file;
   }
   const reader = new window.FileReader();
   // reader.readAsText(file, "UTF-8");

package/src/fastaToJson.js

@@ -62,14 +62,7 @@ function fastaToJson(fileString, options = {}) {
       if (!result) {
         result = createInitialSequence(options);
       }
-      if ("*" === line[line.length - 1]) {
-        //some resultArray are ended with an asterisk
-        parseSequenceLine(line.substring(0, line.length - 1));
-        resultArray.push(result);
-        result = null;
-      } else {
-        parseSequenceLine(line);
-      }
+      parseSequenceLine(line);
     }
     if (options && options.parseFastaAsCircular) {
       result.parsedSequence.circular = true;

package/src/fastqToJson.js

@@ -0,0 +1,80 @@
+import { convertBasePosTraceToPerBpTrace } from "./ab1ToJson.js";
+import splitStringIntoLines from "./utils/splitStringIntoLines.js";
+import validateSequenceArray from "./utils/validateSequenceArray";
+
+/**
+ * parses a fasta file that may or may not contain multiple resultArray
+ * @param  {[string]} fileString   [string respresentation of file contents]
+ * @param  {[function]} onFileParsed [callback for a parsed sequence]
+ * @author Joshua P Nixon
+ */
+
+function validateFastqSet(header, sequence, plusSign, quality) {
+  if (header[0] !== "@") {
+    throw new Error("Invalid FASTQ format: header must start with @");
+  }
+  if (plusSign !== "+") {
+    throw new Error("Invalid FASTQ format: plus sign must be +");
+  }
+  if (quality.length !== sequence.length) {
+    throw new Error(
+      "Invalid FASTQ format: quality and sequence must be the same length"
+    );
+  }
+  if (quality.split("").some(char => char < "!")) {
+    throw new Error("Invalid FASTQ format: quality must be at least !");
+  }
+  if (!/^[acgt]+$/i.test(sequence)) {
+    throw new Error("Invalid FASTQ format: sequence must only contain ACGT");
+  }
+}
+
+function fastqToJson(fileString, options = {}) {
+  options.isProtein = false;
+
+  const lines = splitStringIntoLines(fileString);
+  const resultArray = [];
+  // We could check if the number of lines is divisible by 4,
+  // but maybe the file is not properly terminated.
+  for (let i = 0; i + 3 < lines.length; i += 4) {
+    const header = lines[i];
+    const sequence = lines[i + 1];
+    const plusSign = lines[i + 2];
+    const quality = lines[i + 3];
+
+    validateFastqSet(header, sequence, plusSign, quality);
+
+    const newChromatogramData = convertBasePosTraceToPerBpTrace({
+      aTrace: [],
+      tTrace: [],
+      gTrace: [],
+      cTrace: [],
+      basePos: [],
+      baseCalls: sequence.split(""),
+      baseTraces: sequence.split("").map(() => ({
+        aTrace: [],
+        tTrace: [],
+        gTrace: [],
+        cTrace: []
+      })),
+      qualNums: quality.split("").map(char => char.charCodeAt(0) - 33)
+    });
+
+    const result = {
+      success: true,
+      messages: [],
+      parsedSequence: {
+        name: header.slice(1),
+        sequence: sequence,
+        circular: false,
+        description: "",
+        chromatogramData: newChromatogramData
+      }
+    };
+    resultArray.push(result);
+  }
+
+  return validateSequenceArray(resultArray, options);
+}
+
+export default fastqToJson;

package/src/snapgeneToJson.js

@@ -3,7 +3,7 @@
 
 import bufferpack from "bufferpack";
 import { StringDecoder } from "string_decoder";
-import buffer from "buffer";
+import { Buffer } from "buffer";
 
 import getArrayBufferFromFile from "./utils/getArrayBufferFromFile";
 import createInitialSequence from "./utils/createInitialSequence";
@@ -13,8 +13,6 @@ import { get } from "lodash-es";
 import { XMLParser } from "fast-xml-parser";
 import extractFileExtension from "./utils/extractFileExtension";
 
-const Buffer = buffer.Buffer;
-
 async function snapgeneToJson(fileObj, options = {}) {
   try {
     const returnVal = createInitialSequence(options);

package/utils/unmangleUrls.d.ts

@@ -1,5 +1,2 @@
 export function unmangleUrls(str: any): any;
-export function mangleOrStripUrls(str: any, { mangleUrls, doNotMangleOrStripUrls }?: {
-    mangleUrls: any;
-    doNotMangleOrStripUrls: any;
-}): any;
+export function mangleOrStripUrls(str: any, { mangleUrls, doNotMangleOrStripUrls }?: {}): any;