@sjcrh/proteinpaint-server 2.113.0 → 2.114.0

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@sjcrh/proteinpaint-server",
-  "version": "2.113.0",
+  "version": "2.114.0",
   "type": "module",
   "description": "a genomics visualization tool for exploring a cohort's genotype and phenotype data",
   "main": "src/app.js",
@@ -65,9 +65,9 @@
   },
   "dependencies": {
     "@sjcrh/augen": "2.113.0",
-    "@sjcrh/proteinpaint-rust": "2.112.0",
-    "@sjcrh/proteinpaint-shared": "2.113.0",
-    "@sjcrh/proteinpaint-types": "2.113.0",
+    "@sjcrh/proteinpaint-rust": "2.114.0",
+    "@sjcrh/proteinpaint-shared": "2.114.0",
+    "@sjcrh/proteinpaint-types": "2.114.0",
     "@types/express": "^5.0.0",
     "@types/express-session": "^1.18.1",
     "better-sqlite3": "^9.4.1",

package/routes/termdb.cluster.js

@@ -1,5 +1,6 @@
 import path from "path";
 import run_R from "#src/run_R.js";
+import { run_rust } from "@sjcrh/proteinpaint-rust";
 import { termdbClusterPayload } from "#types/checkers";
 import * as utils from "#src/utils.js";
 import serverconfig from "#src/serverconfig.js";
@@ -223,92 +224,199 @@ async function validate_query_geneExpression(ds, genome) {
   }
   throw "unknown queries.geneExpression.src";
 }
+async function validateHDF5File(filePath) {
+  try {
+    const jsonInput = JSON.stringify({
+      hdf5_file: filePath
+    });
+    const result = await run_rust("validateHDF5", jsonInput);
+    return JSON.parse(result);
+  } catch (error) {
+    console.error(`Error validating file: ${error}`);
+    return {
+      status: "error",
+      message: `Validation error`
+    };
+  }
+}
+async function queryGeneExpression(hdf5_file, geneName) {
+  const jsonInput = JSON.stringify({
+    hdf5_file,
+    gene: geneName
+  });
+  try {
+    const result = await run_rust("readHDF5", jsonInput);
+    if (!result || Object.keys(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 ${geneName}`);
+    throw error;
+  }
+}
 async function validateNative(q, ds, genome) {
-  if (!q.file.startsWith(serverconfig.tpmasterdir))
+  if (q.hdf5File === true) {
     q.file = path.join(serverconfig.tpmasterdir, q.file);
-  if (!q.samples)
     q.samples = [];
-  await utils.validate_tabixfile(q.file);
-  q.nochr = await utils.tabix_is_nochr(q.file, null, genome);
-  q.samples = [];
-  {
-    const lines = await utils.get_header_tabix(q.file);
-    if (!lines[0])
-      throw "header line missing from " + q.file;
-    const l = lines[0].split("	");
-    if (l.slice(0, 4).join("	") != "#chr	start	stop	gene")
-      throw "header line has wrong content for columns 1-4";
-    for (let i = 4; i < l.length; i++) {
-      const id = ds.cohort.termdb.q.sampleName2id(l[i]);
-      if (id == void 0)
-        throw "queries.geneExpression: unknown sample from header: " + l[i];
-      q.samples.push(id);
-    }
-  }
-  q.get = async (param) => {
-    const limitSamples = await mayLimitSamples(param, q.samples, ds);
-    if (limitSamples?.size == 0) {
-      return { term2sample2value: /* @__PURE__ */ new Map(), byTermId: {}, bySampleId: {} };
-    }
-    const bySampleId = {};
-    const samples = q.samples || [];
-    if (limitSamples) {
-      for (const sid of limitSamples) {
-        bySampleId[sid] = { label: ds.cohort.termdb.q.id2sampleName(sid) };
-      }
-    } else {
-      for (const sid of samples) {
-        bySampleId[sid] = { label: ds.cohort.termdb.q.id2sampleName(sid) };
+    await utils.file_is_readable(q.file);
+    try {
+      const vr = await validateHDF5File(q.file);
+      if (vr.status !== "success")
+        throw vr.message;
+      if (!vr.sampleNames?.length)
+        throw "HDF5 file has no samples";
+      for (const sn of vr.sampleNames) {
+        const si = ds.cohort.termdb.q.sampleName2id(sn);
+        if (si == void 0)
+          throw "unknown sample from HDF5: " + sn;
+        q.samples.push(si);
       }
+      console.log(`${ds.label}: HDF5 file validated. Format: ${vr.format}, Samples:`, vr.sampleNames.length);
+    } catch (error) {
+      throw `${ds.label}: Failed to validate HDF5 file: ${error}`;
     }
-    const term2sample2value = /* @__PURE__ */ new Map();
-    for (const geneTerm of param.terms) {
-      if (!geneTerm.gene)
-        continue;
-      if (!geneTerm.chr || !Number.isInteger(geneTerm.start) || !Number.isInteger(geneTerm.stop)) {
-        const re = getResultGene(genome, { input: geneTerm.gene, deep: 1 });
-        if (!re.gmlst || re.gmlst.length == 0) {
-          console.warn("unknown gene:" + geneTerm.gene);
-          continue;
+    q.get = async (param) => {
+      const limitSamples = await mayLimitSamples(param, q.samples, ds);
+      if (limitSamples?.size == 0) {
+        return { term2sample2value: /* @__PURE__ */ new Map(), byTermId: {}, bySampleId: {} };
+      }
+      const bySampleId = {};
+      const samples = q.samples || [];
+      if (limitSamples) {
+        for (const sid of limitSamples) {
+          bySampleId[sid] = { label: ds.cohort.termdb.q.id2sampleName(sid) };
+        }
+      } else {
+        for (const sid of samples) {
+          bySampleId[sid] = { label: ds.cohort.termdb.q.id2sampleName(sid) };
         }
-        const i = re.gmlst.find((i2) => i2.isdefault) || re.gmlst[0];
-        geneTerm.start = i.start;
-        geneTerm.stop = i.stop;
-        geneTerm.chr = i.chr;
       }
-      const s2v = {};
-      if (!geneTerm.chr || !Number.isInteger(geneTerm.start) || !Number.isInteger(geneTerm.stop))
-        throw "missing chr/start/stop";
-      await utils.get_lines_bigfile({
-        args: [
-          q.file,
-          (q.nochr ? geneTerm.chr.replace("chr", "") : geneTerm.chr) + ":" + geneTerm.start + "-" + geneTerm.stop
-        ],
-        callback: (line) => {
-          const l = line.split("	");
-          if (l[3].toLowerCase() != geneTerm.gene.toLowerCase())
-            return;
-          for (let i = 4; i < l.length; i++) {
-            const sampleId = samples[i - 4];
-            if (limitSamples && !limitSamples.has(sampleId))
+      const term2sample2value = /* @__PURE__ */ new Map();
+      const byTermId = {};
+      for (const geneTerm of param.terms) {
+        if (!geneTerm.gene)
+          continue;
+        try {
+          const geneQuery = await queryGeneExpression(q.file, geneTerm.gene);
+          const geneData = JSON.parse(geneQuery);
+          const samplesData = geneData.samples || {};
+          const s2v = {};
+          for (const [sampleName, value] of Object.entries(samplesData)) {
+            const sampleId = ds.cohort.termdb.q.sampleName2id(sampleName);
+            if (!sampleId)
               continue;
-            if (!l[i])
+            if (limitSamples && !limitSamples.has(sampleId))
               continue;
-            const v = Number(l[i]);
-            if (Number.isNaN(v))
-              throw "exp value not number";
-            s2v[sampleId] = v;
+            s2v[sampleId] = Number(value);
+          }
+          console.log(`Gene ${geneTerm.gene} has ${Object.keys(s2v).length} samples with data`);
+          if (Object.keys(s2v).length) {
+            term2sample2value.set(geneTerm.gene, s2v);
           }
+        } catch (error) {
+          console.warn(`Error processing gene ${geneTerm.gene}:`, error);
+          continue;
+        }
+      }
+      if (term2sample2value.size == 0) {
+        throw "No data available for the input " + param.terms?.map((g) => g.gene).join(", ");
+      }
+      return { term2sample2value, byTermId, bySampleId };
+    };
+  } else {
+    if (!q.file.startsWith(serverconfig.tpmasterdir)) {
+      q.file = path.join(serverconfig.tpmasterdir, q.file);
+    }
+    if (!q.samples)
+      q.samples = [];
+    await utils.validate_tabixfile(q.file);
+    q.nochr = await utils.tabix_is_nochr(q.file, null, genome);
+    q.samples = [];
+    {
+      const lines = await utils.get_header_tabix(q.file);
+      if (!lines[0])
+        throw "Header line missing from " + q.file;
+      const l = lines[0].split("	");
+      if (l.slice(0, 4).join("	") != "#chr	start	stop	gene") {
+        throw "Header line has wrong content for columns 1-4";
+      }
+      for (let i = 4; i < l.length; i++) {
+        const id = ds.cohort.termdb.q.sampleName2id(l[i]);
+        if (id == void 0) {
+          throw "queries.geneExpression: unknown sample from header: " + l[i];
         }
-      });
-      if (Object.keys(s2v).length)
-        term2sample2value.set(geneTerm.gene, s2v);
+        q.samples.push(id);
+      }
     }
-    const byTermId = {};
-    if (term2sample2value.size == 0)
-      throw "no data available for the input " + param.terms?.map((g) => g.gene).join(", ");
-    return { term2sample2value, byTermId, bySampleId };
-  };
+    q.get = async (param) => {
+      const limitSamples = await mayLimitSamples(param, q.samples, ds);
+      if (limitSamples?.size == 0) {
+        return { term2sample2value: /* @__PURE__ */ new Map(), byTermId: {}, bySampleId: {} };
+      }
+      const bySampleId = {};
+      const samples = q.samples || [];
+      if (limitSamples) {
+        for (const sid of limitSamples) {
+          bySampleId[sid] = { label: ds.cohort.termdb.q.id2sampleName(sid) };
+        }
+      } else {
+        for (const sid of samples) {
+          bySampleId[sid] = { label: ds.cohort.termdb.q.id2sampleName(sid) };
+        }
+      }
+      const term2sample2value = /* @__PURE__ */ new Map();
+      for (const geneTerm of param.terms) {
+        if (!geneTerm.gene)
+          continue;
+        if (!geneTerm.chr || !Number.isInteger(geneTerm.start) || !Number.isInteger(geneTerm.stop)) {
+          const re = getResultGene(genome, { input: geneTerm.gene, deep: 1 });
+          if (!re.gmlst || re.gmlst.length == 0) {
+            console.warn("Unknown gene:" + geneTerm.gene);
+            continue;
+          }
+          const i = re.gmlst.find((i2) => i2.isdefault) || re.gmlst[0];
+          geneTerm.start = i.start;
+          geneTerm.stop = i.stop;
+          geneTerm.chr = i.chr;
+        }
+        const s2v = {};
+        if (!geneTerm.chr || !Number.isInteger(geneTerm.start) || !Number.isInteger(geneTerm.stop)) {
+          throw "Missing chr/start/stop";
+        }
+        await utils.get_lines_bigfile({
+          args: [
+            q.file,
+            (q.nochr ? geneTerm.chr.replace("chr", "") : geneTerm.chr) + ":" + geneTerm.start + "-" + geneTerm.stop
+          ],
+          callback: (line) => {
+            const l = line.split("	");
+            if (l[3].toLowerCase() != geneTerm.gene.toLowerCase())
+              return;
+            for (let i = 4; i < l.length; i++) {
+              const sampleId = samples[i - 4];
+              if (limitSamples && !limitSamples.has(sampleId))
+                continue;
+              if (!l[i])
+                continue;
+              const v = Number(l[i]);
+              if (Number.isNaN(v))
+                throw "Expression value not number";
+              s2v[sampleId] = v;
+            }
+          }
+        });
+        if (Object.keys(s2v).length) {
+          term2sample2value.set(geneTerm.gene, s2v);
+        }
+      }
+      const byTermId = {};
+      if (term2sample2value.size == 0) {
+        throw "No data available for the input " + param.terms?.map((g) => g.gene).join(", ");
+      }
+      return { term2sample2value, byTermId, bySampleId };
+    };
+  }
 }
 export {
   api,

package/routes/termdb.config.js

@@ -155,6 +155,8 @@ function addNonDictionaryQueries(c, ds, genome) {
       details: q.snvindel.details,
       populations: q.snvindel.populations
     };
+    if (q.snvindel.byisoform?.processTwsInOneQuery)
+      q2.snvindel.byisoform = { processTwsInOneQuery: true };
   }
   if (q.trackLst) {
     q2.trackLst = q.trackLst;