@sjcrh/proteinpaint-server 2.145.0 → 2.145.1-1

package/dataset/termdb.test.js

@@ -303,7 +303,6 @@ function termdb_test_default() {
       geneExpression: {
         src: "native",
         file: "files/hg38/TermdbTest/rnaseq/TermdbTest.fpkm.matrix.new.h5",
-        newformat: true,
         unit: "FPKM"
       },
       ssGSEA: {
@@ -314,7 +313,8 @@ function termdb_test_default() {
       },
       rnaseqGeneCount: {
         storage_type: "HDF5",
-        file: "files/hg38/TermdbTest/rnaseq/TermdbTest.geneCounts.h5"
+        newformat: true,
+        file: "files/hg38/TermdbTest/rnaseq/TermdbTest.geneCounts.new.h5"
       },
       singleCell: {
         samples: {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@sjcrh/proteinpaint-server",
-  "version": "2.145.0",
+  "version": "2.145.1-1",
   "type": "module",
   "description": "a genomics visualization tool for exploring a cohort's genotype and phenotype data",
   "main": "src/app.js",
@@ -62,10 +62,10 @@
   "dependencies": {
     "@sjcrh/augen": "2.143.0",
     "@sjcrh/proteinpaint-python": "2.144.0",
-    "@sjcrh/proteinpaint-r": "2.137.2-0",
-    "@sjcrh/proteinpaint-rust": "2.145.0",
-    "@sjcrh/proteinpaint-shared": "2.143.0",
-    "@sjcrh/proteinpaint-types": "2.145.0",
+    "@sjcrh/proteinpaint-r": "2.145.1-0",
+    "@sjcrh/proteinpaint-rust": "2.145.1-0",
+    "@sjcrh/proteinpaint-shared": "2.145.1-0",
+    "@sjcrh/proteinpaint-types": "2.145.1-1",
     "@types/express": "^5.0.0",
     "@types/express-session": "^1.18.1",
     "better-sqlite3": "^9.4.1",

package/routes/termdb.DE.js

@@ -204,7 +204,12 @@ async function run_DE(param, ds, term_results, term_results2) {
   const sample_size_limit = 8;
   if (group1names.length <= sample_size_limit && group2names.length <= sample_size_limit || param.method == "edgeR" || param.method == "limma") {
     const time12 = (/* @__PURE__ */ new Date()).valueOf();
-    const result2 = JSON.parse(await run_R("edge.R", JSON.stringify(expression_input)));
+    let result2;
+    if (q.newformat) {
+      result2 = JSON.parse(await run_R("edge_newh5.R", JSON.stringify(expression_input)));
+    } else {
+      result2 = JSON.parse(await run_R("edge.R", JSON.stringify(expression_input)));
+    }
     mayLog("Time taken to run edgeR:", formatElapsedTime(Date.now() - time12));
     param.method = "edgeR";
     const ql_imagePath = path.join(serverconfig.cachedir, result2.edgeR_ql_image_name[0]);
@@ -254,15 +259,30 @@ async function validate_query_rnaseqGeneCount(ds) {
     if (ds.queries.rnaseqGeneCount.storage_type == "text") {
       samples = (await get_header_txt(q.file, null)).split("	").slice(4);
     } else if (ds.queries.rnaseqGeneCount.storage_type == "HDF5") {
-      const get_samples_from_hdf5 = {
-        input_file: q.file,
-        data_type: "get_samples"
-      };
-      const time1 = (/* @__PURE__ */ new Date()).valueOf();
-      const result = await run_rust("DEanalysis", JSON.stringify(get_samples_from_hdf5));
-      const time2 = (/* @__PURE__ */ new Date()).valueOf();
-      mayLog("Time taken to query gene expression:", time2 - time1, "ms");
-      samples = result.split(",");
+      if (q.newformat) {
+        const get_samples_from_hdf5 = {
+          hdf5_file: q.file,
+          validate: true
+        };
+        const time1 = (/* @__PURE__ */ new Date()).valueOf();
+        const result = await run_rust("readH5", JSON.stringify(get_samples_from_hdf5));
+        const time2 = (/* @__PURE__ */ new Date()).valueOf();
+        mayLog("Time taken to query gene expression:", time2 - time1, "ms");
+        const vr = JSON.parse(result);
+        if (vr.status !== "success") throw vr.message;
+        if (!Array.isArray(vr.samples)) throw "HDF5 file has no samples, please check file.";
+        samples = vr.samples;
+      } else {
+        const get_samples_from_hdf5 = {
+          input_file: q.file,
+          data_type: "get_samples"
+        };
+        const time1 = (/* @__PURE__ */ new Date()).valueOf();
+        const result = await run_rust("DEanalysis", JSON.stringify(get_samples_from_hdf5));
+        const time2 = (/* @__PURE__ */ new Date()).valueOf();
+        mayLog("Time taken to query gene expression:", time2 - time1, "ms");
+        samples = result.split(",");
+      }
     } else throw "unknown storage type:" + ds.queries.rnaseqGeneCount.storage_type;
     q.allSampleSet = new Set(samples);
     const unknownSamples = [];

package/routes/termdb.boxplot.js

@@ -140,7 +140,7 @@ function setUncomputableValues(values) {
 function parseValues(q, data, sampleType, isLog, overlayTerm, divideTerm) {
   const chart2plot2values = /* @__PURE__ */ new Map();
   const uncomputableValues = {};
-  let absMin = null, absMax = null;
+  let absMin = Infinity, absMax = -Infinity;
   for (const val of Object.values(data.samples)) {
     const value = val[q.tw.$id];
     if (!Number.isFinite(value?.value)) continue;
@@ -175,8 +175,8 @@ function parseValues(q, data, sampleType, isLog, overlayTerm, divideTerm) {
     if (!plot2values.has(plot)) plot2values.set(plot, []);
     const values = plot2values.get(plot);
     values.push(value.value);
-    if (absMin === null || value.value < absMin) absMin = value.value;
-    if (absMax === null || value.value > absMax) absMax = value.value;
+    if (value.value < absMin) absMin = value.value;
+    if (value.value > absMax) absMax = value.value;
   }
   return { absMax, absMin, chart2plot2values, uncomputableValues };
 }

package/routes/termdb.cluster.js

@@ -213,22 +213,6 @@ async function validate_query_geneExpression(ds, genome) {
   }
   throw "unknown queries.geneExpression.src";
 }
-async function queryGeneExpression(hdf5_file, geneNames) {
-  const jsonInput = JSON.stringify({
-    hdf5_file,
-    gene: geneNames
-  });
-  try {
-    const result = await run_rust("readHDF5", jsonInput);
-    if (!result || result.length === 0) {
-      throw new Error("Failed to retrieve expression data: Empty or missing response");
-    }
-    return result;
-  } catch (error) {
-    console.error(`Error querying gene expression for ${geneNames}`);
-    throw error;
-  }
-}
 async function queryHDF5(hdf5_file, query) {
   const jsonInput = JSON.stringify({
     hdf5_file,
@@ -250,34 +234,16 @@ async function validateNative(q, ds) {
   q.samples = [];
   try {
     await utils.file_is_readable(q.file);
-    let tmp;
-    if (q.newformat) {
-      tmp = await run_rust("readH5", JSON.stringify({ hdf5_file: q.file, validate: true }));
-    } else {
-      tmp = await run_rust("validateHDF5", JSON.stringify({ hdf5_file: q.file }));
-    }
+    const tmp = await run_rust("readH5", JSON.stringify({ hdf5_file: q.file, validate: true }));
     const vr = JSON.parse(tmp);
     if (vr.status !== "success") throw vr.message;
-    if (q.newformat) {
-      if (!vr.samples?.length) throw "HDF5 file has no samples, please check file.";
-      for (const sn of vr.samples) {
-        const si = ds.cohort.termdb.q.sampleName2id(sn);
-        if (si == void 0) throw `unknown sample ${sn} from HDF5 ${q.file}`;
-        q.samples.push(si);
-      }
-      console.log(`${ds.label}: geneExpression HDF5 file validated. Format: ${vr.format}, Samples:`, vr.samples.length);
-    } else {
-      if (!vr.sampleNames?.length) throw "HDF5 file has no samples, please check file.";
-      for (const sn of vr.sampleNames) {
-        const si = ds.cohort.termdb.q.sampleName2id(sn);
-        if (si == void 0) throw `unknown sample ${sn} from HDF5 ${q.file}`;
-        q.samples.push(si);
-      }
-      console.log(
-        `${ds.label}: geneExpression HDF5 file validated. Format: ${vr.format}, Samples:`,
-        vr.sampleNames.length
-      );
+    if (!vr.samples?.length) throw "HDF5 file has no samples, please check file.";
+    for (const sn of vr.samples) {
+      const si = ds.cohort.termdb.q.sampleName2id(sn);
+      if (si == void 0) throw `unknown sample ${sn} from HDF5 ${q.file}`;
+      q.samples.push(si);
     }
+    console.log(`${ds.label}: geneExpression HDF5 file validated. Format: ${vr.format}, Samples:`, vr.samples.length);
   } catch (error) {
     throw `${ds.label}: Failed to validate geneExpression HDF5 file: ${error}`;
   }
@@ -310,63 +276,32 @@ async function validateNative(q, ds) {
       return { term2sample2value, byTermId };
     }
     const time1 = Date.now();
-    let geneData;
-    if (q.newformat) {
-      geneData = JSON.parse(await queryHDF5(q.file, geneNames));
-    } else {
-      geneData = JSON.parse(await queryGeneExpression(q.file, geneNames));
-    }
+    const geneData = JSON.parse(await queryHDF5(q.file, geneNames));
     mayLog("Time taken to run gene query:", formatElapsedTime(Date.now() - time1));
-    if (q.newformat) {
-      const genesData = geneData.query_output || {};
-      if (!genesData) throw "No expression data returned from HDF5 query";
-      for (const tw of param.terms) {
-        if (!tw.term.gene) continue;
-        const geneResult = genesData[tw.term.gene];
-        if (!geneResult) {
-          console.warn(`No data found for gene ${tw.term.gene} in the response`);
-          continue;
-        }
-        const samplesData = geneResult.samples || {};
-        const s2v = {};
-        for (const sampleName in samplesData) {
-          const sampleId = ds.cohort.termdb.q.sampleName2id(sampleName);
-          if (!sampleId) continue;
-          if (limitSamples && !limitSamples.has(sampleId)) continue;
-          s2v[sampleId] = samplesData[sampleName];
-        }
-        if (Object.keys(s2v).length) {
-          term2sample2value.set(tw.$id, s2v);
-        }
+    const genesData = geneData.query_output || {};
+    if (!genesData) throw "No expression data returned from HDF5 query";
+    for (const tw of param.terms) {
+      if (!tw.term.gene) continue;
+      const geneResult = genesData[tw.term.gene];
+      if (!geneResult) {
+        console.warn(`No data found for gene ${tw.term.gene} in the response`);
+        continue;
       }
-      if (term2sample2value.size == 0) {
-        throw "No data available for the input " + param.terms?.map((tw) => tw.term.gene).join(", ");
+      const samplesData = geneResult.samples || {};
+      const s2v = {};
+      for (const sampleName in samplesData) {
+        const sampleId = ds.cohort.termdb.q.sampleName2id(sampleName);
+        if (!sampleId) continue;
+        if (limitSamples && !limitSamples.has(sampleId)) continue;
+        s2v[sampleId] = samplesData[sampleName];
       }
-    } else {
-      const genesData = geneData.genes || { [geneNames[0]]: geneData };
-      for (const tw of param.terms) {
-        if (!tw.term.gene) continue;
-        const geneResult = genesData[tw.term.gene];
-        if (!geneResult) {
-          console.warn(`No data found for gene ${tw.term.gene} in the response`);
-          continue;
-        }
-        const samplesData = geneResult.samples || {};
-        const s2v = {};
-        for (const [sampleName, value] of Object.entries(samplesData)) {
-          const sampleId = ds.cohort.termdb.q.sampleName2id(sampleName);
-          if (!sampleId) continue;
-          if (limitSamples && !limitSamples.has(sampleId)) continue;
-          s2v[sampleId] = value;
-        }
-        if (Object.keys(s2v).length) {
-          term2sample2value.set(tw.$id, s2v);
-        }
-      }
-      if (term2sample2value.size == 0) {
-        throw "No data available for the input " + param.terms?.map((tw) => tw.term.gene).join(", ");
+      if (Object.keys(s2v).length) {
+        term2sample2value.set(tw.$id, s2v);
       }
     }
+    if (term2sample2value.size == 0) {
+      throw "No data available for the input " + param.terms?.map((tw) => tw.term.gene).join(", ");
+    }
     return { term2sample2value, byTermId, bySampleId };
   };
 }