@jbrowse/plugin-bed 4.3.0 → 5.0.0-beta.10

package/esm/generateRastairMethylFeature.js

@@ -0,0 +1,38 @@
+import { parseStrand } from './util.js';
+// Rastair (https://www.rastair.com) emits a headered BED/TSV of per-CpG
+// methylation calls from TAPS / mod-C→T data. Its layout differs from modkit
+// bedMethyl, so it's identified by the header column names rather than
+// positionally: #chr start end name beta_est strand unmod mod no_snp snp
+// coverage genotype gt_p_score gt_conf_score
+const RASTAIR_COLUMNS = ['beta_est', 'unmod', 'mod', 'coverage'];
+export function isRastairMethylFeature(names) {
+    return RASTAIR_COLUMNS.every(col => names.includes(col));
+}
+export function generateRastairMethylFeature({ splitLine, names, uniqueId, refName, start, end, }) {
+    const col = (name) => splitLine[names.indexOf(name)];
+    // beta_est is a 0-1 fraction; the shared bedMethyl display path expects a
+    // 0-100 percentage (modkit fraction_modified), so scale to match
+    const fraction_modified = Number(col('beta_est')) * 100;
+    return {
+        uniqueId,
+        refName,
+        start,
+        end,
+        // rastair reports modified C (5mC); use modkit's 'm' code so it groups and
+        // colors as 5mC methylation in the shared bedMethyl display path
+        code: 'm',
+        source: 'm',
+        score: fraction_modified,
+        fraction_modified,
+        strand: parseStrand(col('strand')),
+        n_valid_cov: col('coverage'),
+        n_mod: col('mod'),
+        n_canonical: col('unmod'),
+        beta_est: col('beta_est'),
+        no_snp: col('no_snp'),
+        snp: col('snp'),
+        coverage: col('coverage'),
+        genotype: col('genotype'),
+        gt_conf_score: col('gt_conf_score'),
+    };
+}

package/esm/generateRepeatMaskerFeature.d.ts

@@ -1,51 +1,17 @@
-export declare function isRepeatMaskerDescriptionField(desc?: unknown): desc is string;
-export declare function generateRepeatMaskerFeature({ uniqueId, refName, start, end, description, ...rest }: {
-    uniqueId: string;
-    refName: string;
-    start: number;
-    end: number;
-    description: string;
-    [key: string]: unknown;
-}): {
-    uniqueId: string;
-    refName: string;
-    start: number;
-    end: number;
-    bitsw_score: string | undefined;
-    percent_div: string | undefined;
-    percent_del: string | undefined;
-    percent_ins: string | undefined;
-    query_chr: string | undefined;
-    query_begin: string | undefined;
-    query_end: string | undefined;
-    query_remaining: string | undefined;
-    orientation: string | undefined;
-    matching_repeat_name: string | undefined;
-    matching_repeat_class: string | undefined;
-    matching_repeat_begin: string | undefined;
-    matching_repeat_end: string | undefined;
-    matching_repeat_remaining: string | undefined;
-    repeat_id: string | undefined;
-    description?: undefined;
-} | {
-    uniqueId: string;
-    refName: string;
-    start: number;
-    end: number;
-    description: undefined;
-    bitsw_score?: undefined;
-    percent_div?: undefined;
-    percent_del?: undefined;
-    percent_ins?: undefined;
-    query_chr?: undefined;
-    query_begin?: undefined;
-    query_end?: undefined;
-    query_remaining?: undefined;
-    orientation?: undefined;
-    matching_repeat_name?: undefined;
-    matching_repeat_class?: undefined;
-    matching_repeat_begin?: undefined;
-    matching_repeat_end?: undefined;
-    matching_repeat_remaining?: undefined;
-    repeat_id?: undefined;
-};
+export declare function parseRepeatMaskerDescription(desc: unknown): {
+    bitsw_score: string;
+    percent_div: string;
+    percent_del: string;
+    percent_ins: string;
+    query_chr: string;
+    query_begin: string;
+    query_end: string;
+    query_remaining: string;
+    orientation: string;
+    matching_repeat_name: string;
+    matching_repeat_class: string;
+    matching_repeat_begin: string;
+    matching_repeat_end: string;
+    matching_repeat_remaining: string;
+    repeat_id: string;
+} | undefined;

package/esm/generateRepeatMaskerFeature.js

@@ -1,43 +1,33 @@
-export function isRepeatMaskerDescriptionField(desc) {
+const NUMERIC_PARTS = [0, 1, 2, 3, 5, 6];
+// RepeatMasker .out-derived description: 15 space-separated fields, numeric at
+// positions 0,1,2,3,5,6. A literal rather than an object built from the field
+// list, since this runs on every line of a RepeatMasker track.
+export function parseRepeatMaskerDescription(desc) {
     if (typeof desc !== 'string') {
-        return false;
+        return undefined;
     }
-    else {
-        const ret = desc.trim().split(' ');
-        return [0, 1, 2, 3, 5, 6].every(s => ret[s] !== undefined ? !Number.isNaN(+ret[s]) : false);
+    const p = desc.trim().split(' ');
+    for (const i of NUMERIC_PARTS) {
+        const v = p[i];
+        if (v === undefined || Number.isNaN(+v)) {
+            return undefined;
+        }
     }
-}
-function makeRepeatTrackDescription(description) {
-    if (isRepeatMaskerDescriptionField(description)) {
-        const [bitsw_score, percent_div, percent_del, percent_ins, query_chr, query_begin, query_end, query_remaining, orientation, matching_repeat_name, matching_repeat_class, matching_repeat_begin, matching_repeat_end, matching_repeat_remaining, repeat_id,] = description.trim().split(' ');
-        return {
-            bitsw_score,
-            percent_div,
-            percent_del,
-            percent_ins,
-            query_chr,
-            query_begin,
-            query_end,
-            query_remaining,
-            orientation,
-            matching_repeat_name,
-            matching_repeat_class,
-            matching_repeat_begin,
-            matching_repeat_end,
-            matching_repeat_remaining,
-            repeat_id,
-        };
-    }
-    return { description };
-}
-export function generateRepeatMaskerFeature({ uniqueId, refName, start, end, description, ...rest }) {
-    const { subfeatures, ...rest2 } = rest;
     return {
-        ...rest2,
-        ...makeRepeatTrackDescription(description),
-        uniqueId,
-        refName,
-        start,
-        end,
+        bitsw_score: p[0] ?? '',
+        percent_div: p[1] ?? '',
+        percent_del: p[2] ?? '',
+        percent_ins: p[3] ?? '',
+        query_chr: p[4] ?? '',
+        query_begin: p[5] ?? '',
+        query_end: p[6] ?? '',
+        query_remaining: p[7] ?? '',
+        orientation: p[8] ?? '',
+        matching_repeat_name: p[9] ?? '',
+        matching_repeat_class: p[10] ?? '',
+        matching_repeat_begin: p[11] ?? '',
+        matching_repeat_end: p[12] ?? '',
+        matching_repeat_remaining: p[13] ?? '',
+        repeat_id: p[14] ?? '',
     };
 }

package/esm/generateUcscTranscript.d.ts

@@ -1,12 +1,20 @@
 import type { MinimalFeature } from './types.ts';
-export declare function isUcscTranscript({ thickStart, blockCount, strand, }: {
-    thickStart?: unknown;
-    blockCount?: unknown;
+interface UcscTranscriptCheck {
+    thickStart?: number;
+    thickEnd?: number;
+    blockCount?: number;
     strand?: number;
-}): boolean;
+}
+interface ValidUcscTranscriptCheck {
+    thickStart: number;
+    thickEnd: number;
+    blockCount: number;
+    strand: number;
+}
+export declare function isUcscTranscript(input: UcscTranscriptCheck): input is ValidUcscTranscriptCheck;
 interface UcscTranscriptInput {
     uniqueId: string;
-    strand?: number;
+    strand: number;
     thickStart: number;
     thickEnd: number;
     refName: string;
@@ -15,10 +23,10 @@ interface UcscTranscriptInput {
     subfeatures: MinimalFeature[];
     [key: string]: unknown;
 }
-interface UcscTranscriptOutput {
+interface UcscTranscriptOutput extends MinimalFeature {
     uniqueId: string;
     strand: number;
-    type: string;
+    type: 'mRNA' | 'transcript';
     refName: string;
     start: number;
     end: number;

package/esm/generateUcscTranscript.js

@@ -1,35 +1,99 @@
-export function isUcscTranscript({ thickStart, blockCount, strand, }) {
-    return (typeof thickStart === 'number' &&
-        typeof blockCount === 'number' &&
-        strand !== 0);
+import { copyExcept } from './copyExcept.js';
+// type guard: narrows input fields to required when true
+// strand===0 means unstranded — likely not a gene
+//
+// Blocks are what make a transcript, and UCSC's own format definition calls
+// them exons. Thickness says whether the transcript CODES, which is a separate
+// axis: an lncRNA, a spliced EST and any other 12-column feature with
+// thickStart===thickEnd is a transcript by every structural measure, so it is
+// not tested here. `generateUcscTranscript` reads the same equality to emit
+// exons and no CDS.
+//
+// This does conflate a spliced alignment with a transcript, deliberately: a
+// bigPsl track (`est`, `intronEst`, `mrna`, `xenoMrna`, IsoSeq) has an empty
+// thick range too, so it promotes where the same alignment as GFF3
+// `cDNA_match` does not. BED12 carries no column that separates one from an
+// lncRNA, and the promotion stays honest without one — no CDS children, so the
+// sequence panel withholds CDS and protein and HGVS stays `n.`, and the type
+// is `transcript` rather than `mRNA`.
+export function isUcscTranscript(input) {
+    return (input.thickStart !== undefined &&
+        input.thickEnd !== undefined &&
+        input.blockCount !== undefined &&
+        input.blockCount > 0 &&
+        input.strand !== undefined &&
+        input.strand !== 0);
 }
+// phase = (3 - cumulative_cds_width % 3) % 3, computed in transcriptional order
 function calculatePhasesFromCds(cdsRegions, strand) {
     const sorted = [...cdsRegions].sort((a, b) => strand > 0 ? a.start - b.start : b.start - a.start);
     const phaseMap = new Map();
     let cumulativeWidth = 0;
     for (const cds of sorted) {
-        const phase = (3 - (cumulativeWidth % 3)) % 3;
-        phaseMap.set(cds.start, phase);
+        phaseMap.set(cds.start, (3 - (cumulativeWidth % 3)) % 3);
         cumulativeWidth += cds.end - cds.start;
     }
     return phaseMap;
 }
+// convert UCSC exonFrames (0,1,2) to GFF phase: frame 0→0, frame 1→2, frame 2→1
+// UCSC frame = bases from start of codon; GFF phase = bases to skip to reach next codon
+// https://genome.ucsc.edu/FAQ/FAQformat.html#format1 (bigGenePred exonFrames)
+// https://github.com/The-Sequence-Ontology/Specifications/blob/master/gff3.md (phase)
 function frameToPhase(frame) {
     return (3 - frame) % 3;
 }
+function parseFrames(frames) {
+    return typeof frames === 'string'
+        ? frames.replace(/,$/, '').split(',').map(Number)
+        : frames;
+}
+const TRANSCRIPT_OWN = new Set([
+    'strand',
+    'uniqueId',
+    'start',
+    'end',
+    'thickStart',
+    'thickEnd',
+    'refName',
+    'subfeatures',
+    'chrom',
+    'chromStart',
+    'chromEnd',
+    'chromStarts',
+    'blockStarts',
+    'blockSizes',
+    'blockCount',
+]);
 export function generateUcscTranscript(data) {
-    const { strand = 0, uniqueId, start, end, ...rest } = data;
-    const { subfeatures: oldSubfeatures, thickStart, thickEnd, refName, chrom: _1, chromStart: _2, chromEnd: _3, chromStarts: _4, blockStarts: _5, blockSizes: _6, blockCount: _7, ...rest2 } = rest;
-    const exonFrames = (rest2.exonFrames ?? rest2._exonFrames);
+    const { strand, uniqueId, start, end, thickStart, thickEnd, refName, subfeatures: oldSubfeatures, } = data;
+    const rest = copyExcept(data, TRANSCRIPT_OWN);
+    // exonFrames from bigGenePred - the @gmod/bed parser returns it in genomic order.
+    // _exonFrames fallback supports BED files that use the underscore-prefixed column
+    // name, which the parser leaves as raw text because it is not a standard column.
+    const exonFrames = parseFrames(rest.exonFrames ?? rest._exonFrames);
     const feats = oldSubfeatures
         .filter(child => child.type === 'block')
         .sort((a, b) => a.start - b.start);
     const fiveUTR = strand > 0 ? 'five_prime_UTR' : 'three_prime_UTR';
     const threeUTR = strand > 0 ? 'three_prime_UTR' : 'five_prime_UTR';
-    const { cdsEndStat, cdsStartStat } = rest2;
-    if (cdsStartStat === 'none' && cdsEndStat === 'none') {
+    // thickStart===thickEnd is UCSC for no coding region, and the block walk
+    // below cannot express that: a thick point inside a block would emit a
+    // zero-length CDS carrying a phase, and one outside every block would make
+    // the whole transcript UTR.
+    //
+    // cdsStartStat and cdsEndStat `none` alone do not mean non-coding: UCSC's
+    // GENCODE bigGenePred (gencodeV44.bb through V50) writes `none` on every
+    // transcript, and reading it as non-coding stripped the CDS from all 370,886
+    // of V50's coding ones, TP53 included. exonFrames tells them apart there:
+    // every coding transcript has a block with a frame, every other one has -1
+    // throughout.
+    const { cdsEndStat, cdsStartStat } = rest;
+    if (thickStart === thickEnd ||
+        (cdsStartStat === 'none' &&
+            cdsEndStat === 'none' &&
+            !exonFrames?.some(frame => frame >= 0))) {
         return {
-            ...rest2,
+            ...rest,
             uniqueId,
             strand,
             type: 'transcript',
@@ -39,15 +103,15 @@ export function generateUcscTranscript(data) {
             subfeatures: feats.map(e => ({ ...e, type: 'exon' })),
         };
     }
+    // If exonFrames not available, calculate phases from CDS regions
     let calculatedPhases;
     if (!exonFrames) {
         const cdsRegions = [];
         for (const block of feats) {
-            const { start, end } = block;
-            if (thickStart < end && thickEnd > start) {
+            if (thickStart < block.end && thickEnd > block.start) {
                 cdsRegions.push({
-                    start: Math.max(start, thickStart),
-                    end: Math.min(end, thickEnd),
+                    start: Math.max(block.start, thickStart),
+                    end: Math.min(block.end, thickEnd),
                 });
             }
         }
@@ -55,20 +119,28 @@ export function generateUcscTranscript(data) {
     }
     const subfeatures = [];
     for (const [i, feat] of feats.entries()) {
-        const block = feat;
-        const { start, end } = block;
-        if (thickStart >= end) {
-            subfeatures.push({ type: fiveUTR, start, end, refName });
+        const { start: bStart, end: bEnd } = feat;
+        if (thickStart >= bEnd) {
+            // entire block is 5' UTR
+            subfeatures.push({ type: fiveUTR, start: bStart, end: bEnd, refName });
         }
-        else if (thickEnd <= start) {
-            subfeatures.push({ type: threeUTR, start, end, refName });
+        else if (thickEnd <= bStart) {
+            // entire block is 3' UTR
+            subfeatures.push({ type: threeUTR, start: bStart, end: bEnd, refName });
         }
         else {
-            if (start < thickStart) {
-                subfeatures.push({ type: fiveUTR, start, end: thickStart, refName });
+            // block overlaps CDS region - may have UTR on either side
+            if (bStart < thickStart) {
+                subfeatures.push({
+                    type: fiveUTR,
+                    start: bStart,
+                    end: thickStart,
+                    refName,
+                });
             }
-            const cdsStart = Math.max(start, thickStart);
-            const cdsEnd = Math.min(end, thickEnd);
+            const cdsStart = Math.max(bStart, thickStart);
+            const cdsEnd = Math.min(bEnd, thickEnd);
+            // Get phase from exonFrames (with conversion) or calculated phases
             let phase = 0;
             if (exonFrames) {
                 const frame = exonFrames[i];
@@ -86,13 +158,18 @@ export function generateUcscTranscript(data) {
                 end: cdsEnd,
                 refName,
             });
-            if (end > thickEnd) {
-                subfeatures.push({ type: threeUTR, start: thickEnd, end, refName });
+            if (bEnd > thickEnd) {
+                subfeatures.push({
+                    type: threeUTR,
+                    start: thickEnd,
+                    end: bEnd,
+                    refName,
+                });
             }
         }
     }
     return {
-        ...rest2,
+        ...rest,
         uniqueId,
         strand,
         type: 'mRNA',

package/esm/index.d.ts

@@ -1,5 +1,7 @@
 import Plugin from '@jbrowse/core/Plugin';
 import type PluginManager from '@jbrowse/core/PluginManager';
+export { default as BedTabixAdapter } from './BedTabixAdapter/BedTabixAdapter.ts';
+export { default as bedTabixConfigSchema } from './BedTabixAdapter/configSchema.ts';
 export default class BedPlugin extends Plugin {
     name: string;
     install(pluginManager: PluginManager): void;

package/esm/index.js

@@ -1,20 +1,24 @@
 import Plugin from '@jbrowse/core/Plugin';
-import BedAdapterF from "./BedAdapter/index.js";
-import BedGraphAdapterF from "./BedGraphAdapter/index.js";
-import BedGraphTabixAdapterF from "./BedGraphTabixAdapter/index.js";
-import BedTabixAdapterF from "./BedTabixAdapter/index.js";
-import BedpeAdapterF from "./BedpeAdapter/index.js";
-import BigBedAdapterF from "./BigBedAdapter/index.js";
-import GuessAdapterF from "./GuessAdapter/index.js";
+import BedAdapterF from './BedAdapter/index.js';
+import BedGraphAdapterF from './BedGraphAdapter/index.js';
+import BedGraphTabixAdapterF from './BedGraphTabixAdapter/index.js';
+import BedTabixAdapterF from './BedTabixAdapter/index.js';
+import BedpeAdapterF from './BedpeAdapter/index.js';
+import BigBedAdapterF from './BigBedAdapter/index.js';
+import BigBedTextSearchAdapterF from './BigBedTextSearchAdapter/index.js';
+import StarFusionAdapterF from './StarFusionAdapter/index.js';
+export { default as BedTabixAdapter } from './BedTabixAdapter/BedTabixAdapter.js';
+export { default as bedTabixConfigSchema } from './BedTabixAdapter/configSchema.js';
 export default class BedPlugin extends Plugin {
     name = 'BedPlugin';
     install(pluginManager) {
         BigBedAdapterF(pluginManager);
+        BigBedTextSearchAdapterF(pluginManager);
         BedAdapterF(pluginManager);
         BedpeAdapterF(pluginManager);
+        StarFusionAdapterF(pluginManager);
         BedTabixAdapterF(pluginManager);
         BedGraphAdapterF(pluginManager);
         BedGraphTabixAdapterF(pluginManager);
-        GuessAdapterF(pluginManager);
     }
 }

package/esm/util.d.ts

@@ -1,5 +1,21 @@
+import BED from '@gmod/bed';
 import type { MinimalFeature } from './types.ts';
-import type BED from '@gmod/bed';
+import type { StatusCallback } from '@jbrowse/core/util';
+export interface FeatureData {
+    uniqueId: string;
+    refName: string;
+    start: number;
+    end: number;
+    strand?: number;
+    score?: number;
+    type?: string;
+    subfeatures?: MinimalFeature[];
+    [key: string]: unknown;
+}
+export declare function bucketBedLines(buffer: Uint8Array, statusCallback?: StatusCallback): {
+    header: string;
+    features: Record<string, string[]>;
+};
 export declare function makeBlocks({ start, uniqueId, refName, chromStarts, blockCount, blockSizes, blockStarts, }: {
     blockCount: number;
     start: number;
@@ -8,36 +24,27 @@ export declare function makeBlocks({ start, uniqueId, refName, chromStarts, bloc
     chromStarts?: number[];
     blockSizes?: number[];
     blockStarts?: number[];
-}): {
-    uniqueId: string;
-    start: number;
-    end: number;
-    refName: string;
-    type: string;
-}[];
-export declare function featureData({ line, colRef, colStart, colEnd, scoreColumn, parser, uniqueId, names, disableGeneHeuristic, }: {
-    line: string;
+}): MinimalFeature[];
+export declare function parseNamesFromHeader(header: string): string[] | undefined;
+export declare function resolveColumnNames(columnNames: string[], getHeader: () => Promise<string>): Promise<string[] | undefined>;
+export declare function makeParser({ autoSql, columnNames, }: {
+    autoSql?: string;
+    columnNames?: string[];
+}): BED;
+export declare function bedFeatureLocus({ splitLine, colRef, colStart, colEnd, oneBased, hasEndColumn, }: {
+    splitLine: string[];
     colRef: number;
     colStart: number;
     colEnd: number;
-    scoreColumn: string;
-    parser: BED;
-    uniqueId: string;
-    names?: string[];
-    disableGeneHeuristic?: boolean;
-}): FeatureData;
-interface FeatureData {
-    uniqueId: string;
+    oneBased?: boolean;
+    hasEndColumn?: boolean;
+}): {
     refName: string;
     start: number;
     end: number;
-    strand?: number | string;
-    score?: number;
-    type?: string;
-    subfeatures?: MinimalFeature[];
-    [key: string]: unknown;
-}
-export declare function featureData2({ splitLine, refName, start, end, parser, uniqueId, scoreColumn, names, disableGeneHeuristic, }: {
+};
+export declare function parseStrand(strand: string | number | undefined): number;
+export declare function featureData({ splitLine, refName, start, end, parser, uniqueId, scoreColumn, names, disableGeneHeuristic, }: {
     splitLine: string[];
     refName: string;
     start: number;
@@ -48,5 +55,3 @@ export declare function featureData2({ splitLine, refName, start, end, parser, u
     names?: string[];
     disableGeneHeuristic?: boolean;
 }): FeatureData;
-export declare function arrayify(f?: string | number[]): number[] | undefined;
-export {};