jbrowse-plugin-msaview 5.16.0 → 5.16.2

package/dist/LaunchMsaView/useQueryRowName.js

@@ -19,11 +19,11 @@ import { findQueryRow } from './detectQueryRow';
 export function resolveQueryRowName(queryRow, fallback) {
     return (queryRow.querySeqName || (queryRow.names.includes(fallback) ? fallback : ''));
 }
-export function useQueryRowName(msaText, proteinSequence) {
+export function useQueryRowName(msaText, proteinSequence, preferredName) {
     const [override, setOverride] = useState();
     // parsing runs on every keystroke in the paste box otherwise, and an
     // alignment of a few hundred rows is not free
-    const { names, match } = useMemo(() => findQueryRow(msaText, proteinSequence), [msaText, proteinSequence]);
+    const { names, match } = useMemo(() => findQueryRow(msaText, proteinSequence, preferredName), [msaText, proteinSequence, preferredName]);
     // a row picked from one alignment says nothing about the next one pasted
     const picked = override !== undefined && (names.length === 0 || names.includes(override))
         ? override

package/dist/LaunchMsaView/util.d.ts

@@ -8,6 +8,8 @@ export declare function getTranscriptLength(feature: Feature): {
 };
 export declare function getId(val?: Feature): string;
 export declare function featureMatchesId(feature: Feature, id: string): boolean;
+/** the dataset entry for this transcript, else for the gene it belongs to */
+export declare function findDatasetId(transcript: Feature, gene: Feature, ids: ReadonlySet<string>): string | undefined;
 export declare function getTranscriptDisplayName(val?: Feature): string;
 export declare function getGeneDisplayName(val?: Feature): string;
 export declare function getBlastViewTitle(feature: Feature, transcript: Feature): string;

package/dist/LaunchMsaView/util.js

@@ -39,6 +39,13 @@ export function getTranscriptLength(feature) {
 export function getId(val) {
     return val?.id() ?? '';
 }
+// a proteome downloaded from NCBI or Ensembl names each protein by the CDS's
+// protein_id rather than by its transcript
+function getCdsProteinIds(transcript) {
+    return (transcript.get('subfeatures') ?? [])
+        .filter(f => f.get('type') === 'CDS')
+        .map(f => f.get('protein_id'));
+}
 function getMatchableIds(val) {
     return val
         ? uniqueDefined([
@@ -46,12 +53,43 @@ function getMatchableIds(val) {
             val.get('name'),
             val.get('id'),
             val.get('transcript_id'),
+            val.get('gene_name'),
+            val.get('locus_tag'),
+            ...getCdsProteinIds(val),
         ])
         : [];
 }
 export function featureMatchesId(feature, id) {
     return getMatchableIds(feature).includes(id);
 }
+const geneIdPrefix = /^gene[-:]/;
+// A set keyed by gene, such as an OrthoFinder run over one protein per gene,
+// names the gene by its GFF3 ID without NCBI's `gene-` or Ensembl's `gene:`
+function getGeneMatchableIds(gene) {
+    const id = gene.get('id');
+    return uniqueDefined([
+        id,
+        id?.replace(geneIdPrefix, ''),
+        gene.get('gene_id'),
+        gene.get('locus_tag'),
+    ]);
+}
+// OrthoFinder writes every gene name with each character that Newick or a TSV
+// cannot hold replaced by `_` (its utils/gene_ids.py clean_name), so NCBI's
+// gene `E(spl)m2-BFM` is `E_spl_m2-BFM` in its results. It also replaces control
+// characters, which a GFF3 attribute cannot carry unescaped.
+const orthoFinderUnsafe = /[\s:;(),[\]='"]/g;
+function findIn(candidates, ids) {
+    return (candidates.find(id => ids.has(id)) ??
+        candidates
+            .map(id => id.replace(orthoFinderUnsafe, '_'))
+            .find(id => ids.has(id)));
+}
+/** the dataset entry for this transcript, else for the gene it belongs to */
+export function findDatasetId(transcript, gene, ids) {
+    return (findIn(getMatchableIds(transcript), ids) ??
+        findIn(getGeneMatchableIds(gene), ids));
+}
 export function getTranscriptDisplayName(val) {
     return val ? joinDefined(' ', [val.get('name'), val.get('id')]) : '';
 }

package/dist/MsaViewPanel/displayedRefName.d.ts

@@ -0,0 +1,19 @@
+interface Resolver {
+    initialized?: boolean;
+    getCanonicalRefName: (refName: string) => string | undefined;
+}
+/**
+ * The name a view's displayed regions use for `refName`'s chromosome. core's
+ * bpToPx, and so centerAt, matches region names exactly and does nothing when
+ * the feature says `chr1` and the view displays `1`. The raw name is kept when
+ * a region already carries it, and when no region resolves to the same
+ * chromosome, so a view this cannot place behaves as it did before.
+ */
+export declare function displayedRefName({ displayedRegions, assembly, refName, }: {
+    displayedRegions: readonly {
+        refName: string;
+    }[];
+    assembly?: Resolver;
+    refName: string;
+}): string;
+export {};

package/dist/MsaViewPanel/displayedRefName.js

@@ -0,0 +1,16 @@
+/**
+ * The name a view's displayed regions use for `refName`'s chromosome. core's
+ * bpToPx, and so centerAt, matches region names exactly and does nothing when
+ * the feature says `chr1` and the view displays `1`. The raw name is kept when
+ * a region already carries it, and when no region resolves to the same
+ * chromosome, so a view this cannot place behaves as it did before.
+ */
+export function displayedRefName({ displayedRegions, assembly, refName, }) {
+    const canonical = (name) => assembly?.initialized ? assembly.getCanonicalRefName(name) : undefined;
+    const wanted = canonical(refName);
+    return displayedRegions.some(r => r.refName === refName) ||
+        wanted === undefined
+        ? refName
+        : (displayedRegions.find(r => canonical(r.refName) === wanted)?.refName ??
+            refName);
+}

package/dist/MsaViewPanel/fetchIndexedMsa.js

@@ -1,36 +1,23 @@
-import { BgzfFilehandle } from '@gmod/bgzf-filehandle';
 import { openLocation } from '@jbrowse/core/util/io';
+import { parseMsaIndex, readMsaBlock } from '../utils/msaIndex';
 // Pull one transcript's whole multiple-alignment out of a single bgzip file
 // keyed by NAME. Given the `.fa.gz` uri, the bgzip index (`.gzi`) and name index
-// (`.idx`, a TSV `name<TAB>offset<TAB>length` of each block's uncompressed byte
-// offset + length) are found by suffix. We fetch the small `.idx` once, look up
-// the name, and random-read just that block — already valid FASTA
+// (`.idx`) are found by suffix. We fetch the small `.idx` once, look up the
+// name, and random-read just that block — already valid FASTA
 // (`>hg38\nSEQ\n>panTro4\nSEQ\n...`). One genome-scale alignment serves any gene
 // with no per-gene files or coordinates.
 export async function fetchIndexedMsa({ location, name, }) {
     const open = (uri) => openLocation({ uri, locationType: 'UriLocation' });
-    const idxText = await open(`${location.uri}.idx`).readFile('utf8');
-    const entry = lookup(idxText, name);
-    if (entry) {
-        const fh = new BgzfFilehandle({
+    const want = versionless(name);
+    const entry = parseMsaIndex(await open(`${location.uri}.idx`).readFile('utf8')).find(e => versionless(e.name) === want);
+    return entry
+        ? readMsaBlock({
             filehandle: open(location.uri),
             gziFilehandle: open(`${location.uri}.gzi`),
-        });
-        const bytes = await fh.read(entry.length, entry.offset);
-        return new TextDecoder().decode(bytes).trim();
-    }
-    return undefined;
+            entry,
+        })
+        : undefined;
 }
 // ids are matched versionless, so version drift between the alignment build and
 // the live annotation never breaks the lookup
 const versionless = (s) => s.replace(/\.\d+$/, '');
-function lookup(idxText, name) {
-    const want = versionless(name);
-    for (const line of idxText.split('\n')) {
-        const [id, offset, length] = line.split('\t');
-        if (id && versionless(id) === want) {
-            return { offset: Number(offset), length: Number(length) };
-        }
-    }
-    return undefined;
-}

package/dist/MsaViewPanel/handleMsaClick.test.d.ts

@@ -0,0 +1 @@
+export {};

package/dist/MsaViewPanel/handleMsaClick.test.js

@@ -0,0 +1,59 @@
+import { describe, expect, test, vi } from 'vitest';
+import stateModelFactory from './model';
+const aliases = { chr1: '1' };
+vi.mock('@jbrowse/core/util', async (importOriginal) => ({
+    ...(await importOriginal()),
+    getSession: () => session,
+}));
+const centerAtCalls = [];
+const connectedView = (displayedRefName) => ({
+    id: 'lgv1',
+    type: 'LinearGenomeView',
+    assemblyNames: ['hg38'],
+    displayedRegions: [{ refName: displayedRefName, start: 0, end: 1000 }],
+    centerAt: (coord, refName) => {
+        centerAtCalls.push([coord, refName]);
+    },
+    navTo: vi.fn(),
+});
+let session = {
+    views: [connectedView('1')],
+    hovered: undefined,
+    assemblyManager: {
+        get: () => ({
+            initialized: true,
+            getCanonicalRefName: (refName) => aliases[refName] ?? refName,
+        }),
+    },
+};
+function clickColumn({ displayed, refName, }) {
+    centerAtCalls.length = 0;
+    session = { ...session, views: [connectedView(displayed)] };
+    const model = stateModelFactory().create({
+        type: 'MsaView',
+        id: 'msaview1',
+        connectedViewId: 'lgv1',
+        querySeqName: 'hg38.chr1',
+        mafRegion: { refName, start: 500, end: 510, assemblyName: 'hg38' },
+    });
+    model.setMSA('>hg38.chr1\nACGTACGTAC');
+    model.handleMsaClick(3);
+    return centerAtCalls;
+}
+describe('clicking an alignment column recenters the connected view', () => {
+    test('on the name the view displays when the feature uses an alias', () => {
+        expect(clickColumn({ displayed: '1', refName: 'chr1' })).toEqual([
+            [503, '1'],
+        ]);
+    });
+    test('on the feature name when it is already a displayed region', () => {
+        expect(clickColumn({ displayed: 'chr1', refName: 'chr1' })).toEqual([
+            [503, 'chr1'],
+        ]);
+    });
+    test('on the feature name when no displayed region is the same chromosome', () => {
+        expect(clickColumn({ displayed: '2', refName: 'chr1' })).toEqual([
+            [503, 'chr1'],
+        ]);
+    });
+});

package/dist/MsaViewPanel/model.js

@@ -5,6 +5,7 @@ import { autorun } from 'mobx';
 import { MSAModelF } from 'react-msaview';
 import { autoLoadProteinDomains, launchAwaitsTranscript, launchBlastIfNeeded, launchOrthologsIfNeeded, loadStoredData, observeProteinHighlights, processInit, runCleanup, storeDataToIndexedDB, showDomainsOnGenomeIfAsked, syncGenomeHoverToMsaColumn, } from './afterCreateAutoruns';
 import { connectedDomains } from './connectedDomains';
+import { displayedRefName } from './displayedRefName';
 import { domainTrackConf } from './domainTrack';
 import { isLaunching } from './isLaunching';
 import { msaCoordToGenomeCoord, msaCoordToGenomeRegions, } from './msaCoordToGenomeCoord';
@@ -609,18 +610,19 @@ export default function stateModelFactory() {
          */
         handleMsaClick(coord) {
             const { connectedView, zoomToBaseLevel } = self;
+            const { assemblyManager } = getSession(self);
             const r2 = msaCoordToGenomeCoord({ model: self, coord });
-            if (!r2 || !connectedView) {
-                return;
-            }
-            // Use the genome coord's own refName for both nav paths — it matches the
-            // connected view's displayed regions. Canonicalizing (e.g. "chr17"->"17")
-            // can miss a view whose regions are an alias (same as the bpToPx path).
-            if (zoomToBaseLevel) {
-                connectedView.navTo(r2);
-            }
-            else {
-                connectedView.centerAt(r2.start, r2.refName);
+            if (r2 && connectedView) {
+                if (zoomToBaseLevel) {
+                    connectedView.navTo(r2);
+                }
+                else {
+                    connectedView.centerAt(r2.start, displayedRefName({
+                        displayedRegions: connectedView.displayedRegions,
+                        assembly: assemblyManager.get(connectedView.assemblyNames[0] ?? ''),
+                        refName: r2.refName,
+                    }));
+                }
             }
         },
     }))

package/dist/OrthoFinderMsaAdapter/OrthoFinderMsaAdapter.d.ts

@@ -0,0 +1,22 @@
+import { BaseAdapter } from '@jbrowse/core/data_adapters/BaseAdapter';
+import type { ProteinEntry } from './parseResults';
+/**
+ * Reads the `Results_*` folder of an OrthoFinder run as a dataset: each protein
+ * id it saw names its orthogroup's alignment and gene tree.
+ */
+export default class OrthoFinderMsaAdapter extends BaseAdapter {
+    orthogroupsP: Promise<Map<string, Map<string, ProteinEntry>>> | undefined;
+    resolvedTreesP: Promise<Map<string, string> | 'per-orthogroup'> | undefined;
+    readResultsFile(path: string): Promise<string>;
+    getOrthogroups(): Promise<Map<string, Map<string, ProteinEntry>>>;
+    getResolvedTrees(): Promise<Map<string, string> | "per-orthogroup">;
+    readAlignment(orthogroup: string): Promise<string>;
+    readTree(orthogroup: string): Promise<string | undefined>;
+    speciesFor(assemblyName?: string): Promise<string[]>;
+    getMSAList(assemblyName?: string): Promise<string[]>;
+    getMSA(proteinId: string, assemblyName?: string): Promise<{
+        msa: string;
+        tree: string | undefined;
+        queryRowName: string;
+    }>;
+}

package/dist/OrthoFinderMsaAdapter/OrthoFinderMsaAdapter.js

@@ -0,0 +1,92 @@
+import { BaseAdapter } from '@jbrowse/core/data_adapters/BaseAdapter';
+import { openLocation } from '@jbrowse/core/util/io';
+import { alignmentRowName, isMissingFile, parseOrthogroups, parseResolvedGeneTrees, resultsFile, } from './parseResults';
+/**
+ * Reads the `Results_*` folder of an OrthoFinder run as a dataset: each protein
+ * id it saw names its orthogroup's alignment and gene tree.
+ */
+export default class OrthoFinderMsaAdapter extends BaseAdapter {
+    readResultsFile(path) {
+        return openLocation(resultsFile(this.getConf('resultsLocation'), path), this.pluginManager).readFile('utf8');
+    }
+    getOrthogroups() {
+        this.orthogroupsP ??= this.readResultsFile('Orthogroups/Orthogroups.tsv')
+            .then(parseOrthogroups)
+            .catch((e) => {
+            this.orthogroupsP = undefined;
+            throw e;
+        });
+        return this.orthogroupsP;
+    }
+    getResolvedTrees() {
+        this.resolvedTreesP ??= this.readResultsFile('Resolved_Gene_Trees/Resolved_Gene_Trees.txt').then(parseResolvedGeneTrees, (e) => {
+            if (isMissingFile(e)) {
+                return 'per-orthogroup';
+            }
+            else {
+                this.resolvedTreesP = undefined;
+                throw e;
+            }
+        });
+        return this.resolvedTreesP;
+    }
+    async readAlignment(orthogroup) {
+        try {
+            return await this.readResultsFile(`MultipleSequenceAlignments/${orthogroup}.fa`);
+        }
+        catch (e) {
+            throw isMissingFile(e)
+                ? new Error(`OrthoFinder wrote no alignment for ${orthogroup}. It skips a few small orthogroups, and writes none at all when run with -og, or as OrthoFinder 2 without -M msa`, { cause: e })
+                : e;
+        }
+    }
+    // only orthogroups of four or more genes have a tree
+    async readTree(orthogroup) {
+        const resolvedTrees = await this.getResolvedTrees();
+        return resolvedTrees === 'per-orthogroup'
+            ? this.readResultsFile(`Resolved_Gene_Trees/${orthogroup}_tree.txt`).catch((e) => {
+                if (isMissingFile(e)) {
+                    return undefined;
+                }
+                else {
+                    throw e;
+                }
+            })
+            : resolvedTrees.get(orthogroup);
+    }
+    // the column named after the assembly when there is one, else every column,
+    // for a run whose proteome files are named some other way
+    async speciesFor(assemblyName) {
+        const bySpecies = await this.getOrthogroups();
+        return assemblyName && bySpecies.has(assemblyName)
+            ? [assemblyName]
+            : [...bySpecies.keys()];
+    }
+    async getMSAList(assemblyName) {
+        const bySpecies = await this.getOrthogroups();
+        const species = await this.speciesFor(assemblyName);
+        return [
+            ...new Set(species.flatMap(s => [...(bySpecies.get(s)?.keys() ?? [])])),
+        ];
+    }
+    async getMSA(proteinId, assemblyName) {
+        const bySpecies = await this.getOrthogroups();
+        const found = (await this.speciesFor(assemblyName))
+            .map(species => ({
+            species,
+            entry: bySpecies.get(species)?.get(proteinId),
+        }))
+            .find(({ entry }) => entry);
+        if (found?.entry) {
+            const { species, entry } = found;
+            const [msa, tree] = await Promise.all([
+                this.readAlignment(entry.orthogroup),
+                this.readTree(entry.orthogroup),
+            ]);
+            return { msa, tree, queryRowName: alignmentRowName(species, entry.id) };
+        }
+        else {
+            throw new Error(`${proteinId} is in no OrthoFinder orthogroup`);
+        }
+    }
+}

package/dist/OrthoFinderMsaAdapter/configSchema.d.ts

@@ -0,0 +1,12 @@
+export default function configSchemaF(): import("@jbrowse/core/configuration/configurationSchema").ConfigurationSchemaType<{
+    readonly resultsLocation: {
+        type: "fileLocation";
+    } & {
+        defaultValue: unknown;
+    };
+    type: {
+        type: "string";
+    } & {
+        defaultValue: string;
+    };
+}, import("@jbrowse/core/configuration/configurationSchema").ConfigurationSchemaOptions<undefined, undefined, true, import("@jbrowse/core/configuration").ConfigurationSchemaRequirement<Partial<Record<string, string[]>>, string>, any>>;

package/dist/OrthoFinderMsaAdapter/configSchema.js

@@ -0,0 +1,13 @@
+import { ConfigurationSchema } from '@jbrowse/core/configuration';
+export default function configSchemaF() {
+    return ConfigurationSchema('OrthoFinderMsaAdapter', {
+        resultsLocation: {
+            type: 'fileLocation',
+            description: 'the Results_* folder of an OrthoFinder run, holding Orthogroups/, MultipleSequenceAlignments/ and Resolved_Gene_Trees/',
+            defaultValue: {
+                uri: '/path/to/Results',
+                locationType: 'UriLocation',
+            },
+        },
+    }, { explicitlyTyped: true });
+}

package/dist/OrthoFinderMsaAdapter/index.d.ts

@@ -0,0 +1,2 @@
+import type PluginManager from '@jbrowse/core/PluginManager';
+export default function OrthoFinderMsaAdapterF(pluginManager: PluginManager): PluginManager;

package/dist/OrthoFinderMsaAdapter/index.js

@@ -0,0 +1,14 @@
+import { AdapterType } from '@jbrowse/core/pluggableElementTypes';
+import configSchemaF from './configSchema';
+export default function OrthoFinderMsaAdapterF(pluginManager) {
+    return pluginManager.addAdapterType(() => {
+        return new AdapterType({
+            name: 'OrthoFinderMsaAdapter',
+            configSchema: configSchemaF(),
+            adapterMetadata: {
+                hiddenFromGUI: true,
+            },
+            getAdapterClass: () => import('./OrthoFinderMsaAdapter').then(t => t.default),
+        });
+    });
+}

package/dist/OrthoFinderMsaAdapter/parseResults.d.ts

@@ -0,0 +1,31 @@
+import type { FileLocation } from '@jbrowse/core/util/types';
+export interface ProteinEntry {
+    orthogroup: string;
+    /** the id as OrthoFinder wrote it, which names the protein's alignment row */
+    id: string;
+}
+/**
+ * OrthoFinder's `Orthogroups/Orthogroups.tsv`, one map per species column from
+ * protein id to its orthogroup. Per species because OrthoFinder leaves a name
+ * two species share unprefixed there: human's `ADAM32` and chicken's are
+ * different genes in different orthogroups. Each id is also listed without a
+ * trailing `.N` version, since a proteome FASTA usually versions its ids where
+ * the matching GFF does not (Ensembl's `ENSP00000269305.4` against
+ * `protein_id=ENSP00000269305`).
+ */
+export declare function parseOrthogroups(tsv: string): Map<string, Map<string, ProteinEntry>>;
+/**
+ * The alignment and tree name each protein `<species>_<id>`, the species with
+ * `.` and space made `_` (OrthoFinder's species_prefixes).
+ */
+export declare function alignmentRowName(species: string, id: string): string;
+/** a file the results folder does not hold, as opposed to one it failed to serve */
+export declare function isMissingFile(e: unknown): boolean;
+/**
+ * OrthoFinder 3 writes every gene tree into one
+ * `Resolved_Gene_Trees/Resolved_Gene_Trees.txt`, a line per orthogroup of the
+ * form `OG0000055: (newick);`
+ */
+export declare function parseResolvedGeneTrees(text: string): Map<string, string>;
+/** a file inside the results folder, addressed the same way the folder is */
+export declare function resultsFile(results: FileLocation, path: string): FileLocation;

package/dist/OrthoFinderMsaAdapter/parseResults.js

@@ -0,0 +1,81 @@
+import { isLocalPathLocation, isUriLocation } from '@jbrowse/core/util/types';
+const versionSuffix = /\.\d+$/;
+/**
+ * OrthoFinder's `Orthogroups/Orthogroups.tsv`, one map per species column from
+ * protein id to its orthogroup. Per species because OrthoFinder leaves a name
+ * two species share unprefixed there: human's `ADAM32` and chicken's are
+ * different genes in different orthogroups. Each id is also listed without a
+ * trailing `.N` version, since a proteome FASTA usually versions its ids where
+ * the matching GFF does not (Ensembl's `ENSP00000269305.4` against
+ * `protein_id=ENSP00000269305`).
+ */
+export function parseOrthogroups(tsv) {
+    const [header = '', ...rows] = tsv.split(/\r?\n/);
+    const species = header.split('\t').slice(1);
+    const bySpecies = new Map(species.map(name => [name, new Map()]));
+    for (const row of rows) {
+        const [orthogroup, ...cells] = row.split('\t');
+        if (orthogroup) {
+            cells.forEach((cell, i) => {
+                const proteins = bySpecies.get(species[i] ?? '');
+                for (const geneId of cell.split(',')) {
+                    const id = geneId.trim();
+                    if (proteins && id) {
+                        const entry = { orthogroup, id };
+                        proteins.set(id, entry);
+                        const unversioned = id.replace(versionSuffix, '');
+                        if (!proteins.has(unversioned)) {
+                            proteins.set(unversioned, entry);
+                        }
+                    }
+                }
+            });
+        }
+    }
+    return bySpecies;
+}
+/**
+ * The alignment and tree name each protein `<species>_<id>`, the species with
+ * `.` and space made `_` (OrthoFinder's species_prefixes).
+ */
+export function alignmentRowName(species, id) {
+    return `${species.replace(/[. ]/g, '_')}_${id}`;
+}
+/** a file the results folder does not hold, as opposed to one it failed to serve */
+export function isMissingFile(e) {
+    return e instanceof Error && /HTTP 40[34]\b|ENOENT/.test(e.message);
+}
+/**
+ * OrthoFinder 3 writes every gene tree into one
+ * `Resolved_Gene_Trees/Resolved_Gene_Trees.txt`, a line per orthogroup of the
+ * form `OG0000055: (newick);`
+ */
+export function parseResolvedGeneTrees(text) {
+    const trees = new Map();
+    for (const line of text.split(/\r?\n/)) {
+        const separator = line.indexOf(': ');
+        if (separator > 0) {
+            trees.set(line.slice(0, separator), line.slice(separator + 2).trim());
+        }
+    }
+    return trees;
+}
+/** a file inside the results folder, addressed the same way the folder is */
+export function resultsFile(results, path) {
+    if (isLocalPathLocation(results)) {
+        return {
+            ...results,
+            localPath: `${results.localPath.replace(/\/+$/, '')}/${path}`,
+        };
+    }
+    else if (isUriLocation(results)) {
+        const folder = new URL(results.uri, results.baseUri).href;
+        return {
+            ...results,
+            uri: new URL(path, folder.endsWith('/') ? folder : `${folder}/`).href,
+        };
+    }
+    else {
+        throw new Error('An OrthoFinder results folder must be given as a URL or a local path');
+    }
+}

package/dist/OrthoFinderMsaAdapter/parseResults.test.d.ts

@@ -0,0 +1 @@
+export {};

package/dist/OrthoFinderMsaAdapter/parseResults.test.js

@@ -0,0 +1,47 @@
+import { expect, test } from 'vitest';
+import { alignmentRowName, isMissingFile, parseOrthogroups, parseResolvedGeneTrees, resultsFile, } from './parseResults';
+test('a versioned protein id also answers to its unversioned form', () => {
+    const human = parseOrthogroups('Orthogroup\tHuman\tMouse\nOG0000001\tENSP00000269305.4\tENSMUSP00000104298.3, ENSMUSP00000005371\n').get('Human');
+    expect(human?.get('ENSP00000269305.4')?.orthogroup).toBe('OG0000001');
+    expect(human?.get('ENSP00000269305')).toEqual({
+        orthogroup: 'OG0000001',
+        id: 'ENSP00000269305.4',
+    });
+});
+test('an unversioned alias never displaces a real id of the same name', () => {
+    const groups = parseOrthogroups('Orthogroup\tA\nOG0000001\tP1.2\nOG0000002\tP1\n');
+    expect(groups.get('A')?.get('P1')?.orthogroup).toBe('OG0000002');
+    const reversed = parseOrthogroups('Orthogroup\tA\nOG0000002\tP1\nOG0000001\tP1.2\n');
+    expect(reversed.get('A')?.get('P1')?.orthogroup).toBe('OG0000002');
+});
+test("a name two species share stays each species' own", () => {
+    const groups = parseOrthogroups('Orthogroup\tchicken\thuman\r\nOG0013180\t\tADAM32\r\nOG0015286\tADAM32\t\r\n');
+    expect(groups.get('human')?.get('ADAM32')?.orthogroup).toBe('OG0013180');
+    expect(groups.get('chicken')?.get('ADAM32')?.orthogroup).toBe('OG0015286');
+});
+test("an alignment row is the species, made OrthoFinder's prefix, and the id", () => {
+    expect(alignmentRowName('melanogaster', 'Dmel_CG2671')).toBe('melanogaster_Dmel_CG2671');
+    expect(alignmentRowName('Homo sapiens.v2', 'P1')).toBe('Homo_sapiens_v2_P1');
+});
+test('only a 403, 404 or absent local file reads as missing', () => {
+    expect(isMissingFile(new Error('HTTP 404 fetching https://x/a'))).toBe(true);
+    expect(isMissingFile(new Error('HTTP 403 fetching https://x/a'))).toBe(true);
+    expect(isMissingFile(new Error("ENOENT: no such file, open '/a'"))).toBe(true);
+    expect(isMissingFile(new Error('HTTP 503 fetching https://x/a'))).toBe(false);
+    expect(isMissingFile(new TypeError('Failed to fetch'))).toBe(false);
+});
+test('Resolved_Gene_Trees.txt splits on the first colon-space only', () => {
+    const trees = parseResolvedGeneTrees('OG0000001: (A_x:0.1,B_y:0.2)n0;\nOG0000002: (C:1,D:2);\n');
+    expect(trees.get('OG0000001')).toBe('(A_x:0.1,B_y:0.2)n0;');
+    expect(trees.size).toBe(2);
+});
+test('a results file resolves inside the folder however the folder is written', () => {
+    const file = 'Orthogroups/Orthogroups.tsv';
+    expect(resultsFile({ uri: 'https://x.org/run/Results_Nov05', locationType: 'UriLocation' }, file)).toMatchObject({ uri: `https://x.org/run/Results_Nov05/${file}` });
+    expect(resultsFile({
+        uri: 'Results_Nov05/',
+        baseUri: 'https://x.org/run/config.json',
+        locationType: 'UriLocation',
+    }, file)).toMatchObject({ uri: `https://x.org/run/Results_Nov05/${file}` });
+    expect(resultsFile({ localPath: '/data/Results_Nov05/', locationType: 'LocalPathLocation' }, file)).toMatchObject({ localPath: `/data/Results_Nov05/${file}` });
+});

package/dist/chunks/BgzipFastaMsaAdapter-Y3LYQRQL.js

@@ -0,0 +1,6 @@
+import{a as V,b as q}from"./chunk-WE35XW3J.js";import{a as B}from"./chunk-3G2F2Y7O.js";import{a as v}from"./chunk-YN7VVNZR.js";import{c as k}from"./chunk-QBRNXRNU.js";import"./chunk-KVMUXFPB.js";var T=function(t,r){return T=Object.setPrototypeOf||{__proto__:[]}instanceof Array&&function(e,o){e.__proto__=o}||function(e,o){for(var n in o)Object.prototype.hasOwnProperty.call(o,n)&&(e[n]=o[n])},T(t,r)};function b(t,r){if(typeof r!="function"&&r!==null)throw new TypeError("Class extends value "+String(r)+" is not a constructor or null");T(t,r);function e(){this.constructor=t}t.prototype=r===null?Object.create(r):(e.prototype=r.prototype,new e)}function j(t){var r=typeof Symbol=="function"&&Symbol.iterator,e=r&&t[r],o=0;if(e)return e.call(t);if(t&&typeof t.length=="number")return{next:function(){return t&&o>=t.length&&(t=void 0),{value:t&&t[o++],done:!t}}};throw new TypeError(r?"Object is not iterable.":"Symbol.iterator is not defined.")}function d(t,r){var e=typeof Symbol=="function"&&t[Symbol.iterator];if(!e)return t;var o=e.call(t),n,i=[],a;try{for(;(r===void 0||r-- >0)&&!(n=o.next()).done;)i.push(n.value)}catch(f){a={error:f}}finally{try{n&&!n.done&&(e=o.return)&&e.call(o)}finally{if(a)throw a.error}}return i}function h(t,r,e){if(e||arguments.length===2)for(var o=0,n=r.length,i;o<n;o++)(i||!(o in r))&&(i||(i=Array.prototype.slice.call(r,0,o)),i[o]=r[o]);return t.concat(i||Array.prototype.slice.call(r))}function p(t){return typeof t=="function"}function _(t){var r=function(o){Error.call(o),o.stack=new Error().stack},e=t(r);return e.prototype=Object.create(Error.prototype),e.prototype.constructor=e,e}var E=_(function(t){return function(e){t(this),this.message=e?e.length+` errors occurred during unsubscription:
+`+e.map(function(o,n){return n+1+") "+o.toString()}).join(`
+  `):"",this.name="UnsubscriptionError",this.errors=e}});function A(t,r){if(t){var e=t.indexOf(r);0<=e&&t.splice(e,1)}}var S=(function(){function t(r){this.initialTeardown=r,this.closed=!1,this._parentage=null,this._finalizers=null}return t.prototype.unsubscribe=function(){var r,e,o,n,i;if(!this.closed){this.closed=!0;var a=this._parentage;if(a)if(this._parentage=null,Array.isArray(a))try{for(var f=j(a),s=f.next();!s.done;s=f.next()){var c=s.value;c.remove(this)}}catch(u){r={error:u}}finally{try{s&&!s.done&&(e=f.return)&&e.call(f)}finally{if(r)throw r.error}}else a.remove(this);var l=this.initialTeardown;if(p(l))try{l()}catch(u){i=u instanceof E?u.errors:[u]}var w=this._finalizers;if(w){this._finalizers=null;try{for(var g=j(w),y=g.next();!y.done;y=g.next()){var er=y.value;try{$(er)}catch(u){i=i??[],u instanceof E?i=h(h([],d(i)),d(u.errors)):i.push(u)}}}catch(u){o={error:u}}finally{try{y&&!y.done&&(n=g.return)&&n.call(g)}finally{if(o)throw o.error}}}if(i)throw new E(i)}},t.prototype.add=function(r){var e;if(r&&r!==this)if(this.closed)$(r);else{if(r instanceof t){if(r.closed||r._hasParent(this))return;r._addParent(this)}(this._finalizers=(e=this._finalizers)!==null&&e!==void 0?e:[]).push(r)}},t.prototype._hasParent=function(r){var e=this._parentage;return e===r||Array.isArray(e)&&e.includes(r)},t.prototype._addParent=function(r){var e=this._parentage;this._parentage=Array.isArray(e)?(e.push(r),e):e?[e,r]:r},t.prototype._removeParent=function(r){var e=this._parentage;e===r?this._parentage=null:Array.isArray(e)&&A(e,r)},t.prototype.remove=function(r){var e=this._finalizers;e&&A(e,r),r instanceof t&&r._removeParent(this)},t.EMPTY=(function(){var r=new t;return r.closed=!0,r})(),t})();var gr=S.EMPTY;function z(t){return t instanceof S||t&&"closed"in t&&p(t.remove)&&p(t.add)&&p(t.unsubscribe)}function $(t){p(t)?t():t.unsubscribe()}var m={onUnhandledError:null,onStoppedNotification:null,Promise:void 0,useDeprecatedSynchronousErrorHandling:!1,useDeprecatedNextContext:!1};var x={setTimeout:function(t,r){for(var e=[],o=2;o<arguments.length;o++)e[o-2]=arguments[o];var n=x.delegate;return n?.setTimeout?n.setTimeout.apply(n,h([t,r],d(e))):setTimeout.apply(void 0,h([t,r],d(e)))},clearTimeout:function(t){var r=x.delegate;return(r?.clearTimeout||clearTimeout)(t)},delegate:void 0};function H(t){x.setTimeout(function(){var r=m.onUnhandledError;if(r)r(t);else throw t})}function M(){}var Y=(function(){return I("C",void 0,void 0)})();function K(t){return I("E",void 0,t)}function G(t){return I("N",t,void 0)}function I(t,r,e){return{kind:t,value:r,error:e}}var R=null;function J(t){m.useDeprecatedSynchronousErrorHandling&&R&&(R.errorThrown=!0,R.error=t)}var L=(function(t){b(r,t);function r(e){var o=t.call(this)||this;return o.isStopped=!1,e?(o.destination=e,z(e)&&e.add(o)):o.destination=ir,o}return r.create=function(e,o,n){return new U(e,o,n)},r.prototype.next=function(e){this.isStopped?F(G(e),this):this._next(e)},r.prototype.error=function(e){this.isStopped?F(K(e),this):(this.isStopped=!0,this._error(e))},r.prototype.complete=function(){this.isStopped?F(Y,this):(this.isStopped=!0,this._complete())},r.prototype.unsubscribe=function(){this.closed||(this.isStopped=!0,t.prototype.unsubscribe.call(this),this.destination=null)},r.prototype._next=function(e){this.destination.next(e)},r.prototype._error=function(e){try{this.destination.error(e)}finally{this.unsubscribe()}},r.prototype._complete=function(){try{this.destination.complete()}finally{this.unsubscribe()}},r})(S);var tr=Function.prototype.bind;function C(t,r){return tr.call(t,r)}var or=(function(){function t(r){this.partialObserver=r}return t.prototype.next=function(r){var e=this.partialObserver;if(e.next)try{e.next(r)}catch(o){P(o)}},t.prototype.error=function(r){var e=this.partialObserver;if(e.error)try{e.error(r)}catch(o){P(o)}else P(r)},t.prototype.complete=function(){var r=this.partialObserver;if(r.complete)try{r.complete()}catch(e){P(e)}},t})(),U=(function(t){b(r,t);function r(e,o,n){var i=t.call(this)||this,a;if(p(e)||!e)a={next:e??void 0,error:o??void 0,complete:n??void 0};else{var f;i&&m.useDeprecatedNextContext?(f=Object.create(e),f.unsubscribe=function(){return i.unsubscribe()},a={next:e.next&&C(e.next,f),error:e.error&&C(e.error,f),complete:e.complete&&C(e.complete,f)}):a=e}return i.destination=new or(a),i}return r})(L);function P(t){m.useDeprecatedSynchronousErrorHandling?J(t):H(t)}function nr(t){throw t}function F(t,r){var e=m.onStoppedNotification;e&&x.setTimeout(function(){return e(t,r)})}var ir={closed:!0,next:M,error:nr,complete:M};function ar(t){return p(t?.lift)}function O(t){return function(r){if(ar(r))return r.lift(function(e){try{return t(e,this)}catch(o){this.error(o)}});throw new TypeError("Unable to lift unknown Observable type")}}function Q(t,r,e,o,n){return new fr(t,r,e,o,n)}var fr=(function(t){b(r,t);function r(e,o,n,i,a,f){var s=t.call(this,e)||this;return s.onFinalize=a,s.shouldUnsubscribe=f,s._next=o?function(c){try{o(c)}catch(l){e.error(l)}}:t.prototype._next,s._error=i?function(c){try{i(c)}catch(l){e.error(l)}finally{this.unsubscribe()}}:t.prototype._error,s._complete=n?function(){try{n()}catch(c){e.error(c)}finally{this.unsubscribe()}}:t.prototype._complete,s}return r.prototype.unsubscribe=function(){var e;if(!this.shouldUnsubscribe||this.shouldUnsubscribe()){var o=this.closed;t.prototype.unsubscribe.call(this),!o&&((e=this.onFinalize)===null||e===void 0||e.call(this))}},r})(L);var X=_(function(t){return function(){t(this),this.name="EmptyError",this.message="no elements in sequence"}});function D(t,r){var e=typeof r=="object";return new Promise(function(o,n){var i=new U({next:function(a){o(a),i.unsubscribe()},error:n,complete:function(){e?o(r.defaultValue):n(new X)}});t.subscribe(i)})}function Z(t,r,e,o,n){return function(i,a){var f=e,s=r,c=0;i.subscribe(Q(a,function(l){var w=c++;s=f?t(s,l,w):(f=!0,l),o&&a.next(s)},n&&(function(){f&&a.next(s),a.complete()})))}}function rr(t,r){return O(Z(t,r,arguments.length>=2,!1,!0))}var sr=function(t,r){return t.push(r),t};function N(){return O(function(t,r){rr(sr,[])(t).subscribe(r)})}var W=class extends B{async configurePre(){let r=this.getSubAdapter;if(r)return(await r({...k(this.config),type:"BgzipFastaAdapter"})).dataAdapter;throw new Error("no get subadapter")}configure(){return this.configureP??=this.configurePre().catch(r=>{throw this.configureP=void 0,r}),this.configureP}async readMsaIndex(r){let e=await v(r,this.pluginManager).readFile("utf8");return new Map(V(e).map(o=>[o.name,o]))}async loadMsaIndex(){let r=this.getConf("msaIndexLocation"),e=await this.configure(),o=ur(e.getConf("fastaLocation"),".idx");return"uri"in r&&!r.uri?o?this.readMsaIndex(o).catch(()=>{}):void 0:this.readMsaIndex(r)}getMsaIndex(){return this.msaIndexP??=this.loadMsaIndex().catch(r=>{throw this.msaIndexP=void 0,r}),this.msaIndexP}async groupRows(){let r=await this.configure(),e=new RegExp(this.getConf("msaRegex")),o=new Map;for(let n of await r.getRefNames()){let i=n.split(e)[0],a=o.get(i);a?a.push(n):o.set(i,[n])}return o}getMSARows(){return this.msaRowsP??=this.groupRows().catch(r=>{throw this.msaRowsP=void 0,r}),this.msaRowsP}async getMSAList(){return[...(await this.getMsaIndex()??await this.getMSARows()).keys()]}async readIndexedMSA(r,e){let o=await this.configure(),n=r.get(e);return n?q({filehandle:v(o.getConf("fastaLocation"),this.pluginManager),gziFilehandle:v(o.getConf("gziLocation"),this.pluginManager),entry:n}):""}async readRowsMSA(r){let e=await this.configure(),o=(await this.getMSARows()).get(r)??[];return(await D(e.getFeaturesInMultipleRegions(o.map(i=>({refName:i,start:0,end:1e9,assemblyName:""}))).pipe(N()))).map(i=>`>${i.get("refName")}
+${i.get("seq")}`).join(`
+`)}async getMSA(r){let e=await this.getMsaIndex();return{msa:await(e?this.readIndexedMSA(e,r):this.readRowsMSA(r))}}};function ur(t,r){return"uri"in t?{...t,uri:`${t.uri}${r}`}:"localPath"in t?{...t,localPath:`${t.localPath}${r}`}:void 0}export{W as default};
+//# sourceMappingURL=BgzipFastaMsaAdapter-Y3LYQRQL.js.map