@apollo-annotation/shared 0.3.7 → 0.3.9

package/src/Checks/CDSCheck.ts

@@ -6,6 +6,8 @@ import {
 import { intersection2 } from '@jbrowse/core/util'
 import ObjectID from 'bson-objectid'
 
+import { getPrintableId } from './util'
+
 enum STOP_CODONS {
   'TAG',
   'TAA',
@@ -22,7 +24,6 @@ enum CAUSES {
   'InternalStopCodon',
   'MissingStartCodon',
   'MissingStopCodon',
-  'MultipleOfThree',
 }
 
 const iupacComplements: Record<string, string | undefined> = {
@@ -156,6 +157,10 @@ async function checkMRNA(
   for (const cdsLocation of cdsLocations) {
     const sequence = await getCDSSequence(cdsLocation, strand, getSequence)
     const codons = splitSequenceInCodons(sequence)
+    const cdsEnd =
+      strand === -1
+        ? cdsLocation.at(0)?.min ?? min
+        : cdsLocation.at(-1)?.max ?? max
     if (sequence.length % 3 === 0) {
       const start_codon = codons.at(0)
       if (start_codon && !(start_codon.toUpperCase() in START_CODONS)) {
@@ -172,15 +177,12 @@ async function checkMRNA(
           refSeq: refSeq.toString(),
           start: cdsStart,
           end: cdsStart,
-          message: `Unexpected start codon in feature "${_id}": ${start_codon}`,
+          message: `Unexpected start codon "${start_codon}" in feature "${getPrintableId(feature)}": `,
         })
       }
+
       const lastCodon = codons.at(-1) // Last codon is supposed to be a stop
       if (lastCodon && !(lastCodon.toUpperCase() in STOP_CODONS)) {
-        const cdsEnd =
-          strand === -1
-            ? cdsLocation.at(0)?.min ?? min
-            : cdsLocation.at(-1)?.max ?? max
         checkResults.push({
           _id: new ObjectID().toHexString(),
           name: CHECK_NAME,
@@ -189,19 +191,19 @@ async function checkMRNA(
           refSeq: refSeq.toString(),
           start: cdsEnd,
           end: cdsEnd,
-          message: `Missing stop codon for feature "${_id}"`,
+          message: `Missing stop codon in feature "${getPrintableId(feature)}"`,
         })
       }
     } else {
       checkResults.push({
         _id: new ObjectID().toHexString(),
         name: CHECK_NAME,
-        cause: CAUSES[CAUSES.MultipleOfThree],
+        cause: CAUSES[CAUSES.MissingStopCodon],
         ids,
         refSeq: refSeq.toString(),
-        start: min,
-        end: max,
-        message: `The coding sequence for feature "${_id}" is not a multiple of three`,
+        start: cdsEnd,
+        end: cdsEnd,
+        message: `Missing stop codon in feature "${getPrintableId(feature)}"`,
       })
     }
     for (const [idx, codon] of codons.entries()) {
@@ -220,7 +222,7 @@ async function checkMRNA(
           refSeq: refSeq.toString(),
           start: codonStart,
           end: codonEnd,
-          message: `The coding sequence for feature "${_id}" has an internal stop codon`,
+          message: `Internal stop codon in feature "${getPrintableId(feature)}"`,
         })
       }
     }
@@ -285,7 +287,7 @@ export class CDSCheck extends Check {
   name = CHECK_NAME
   causes = getCauses()
   version = 1
-  default = true
+  isDefault = true
 
   async checkFeature(
     feature: AnnotationFeatureSnapshot,

package/src/Checks/TranscriptCheck.ts

@@ -0,0 +1,138 @@
+import { Check } from '@apollo-annotation/common'
+import {
+  type AnnotationFeatureSnapshot,
+  type CheckResultSnapshot,
+} from '@apollo-annotation/mst'
+import { revcom } from '@jbrowse/core/util'
+import ObjectID from 'bson-objectid'
+
+import { getPrintableId } from './util'
+
+interface SpliceSequence {
+  fivePrimeSeq: string
+  fivePrimeMin: number
+  threePrimeSeq: string
+  threePrimeMin: number
+}
+
+enum CAUSES {
+  'NonCanonicalSpliceSiteAtFivePrime',
+  'NonCanonicalSpliceSiteAtThreePrime',
+}
+const CHECK_NAME = 'TranscriptCheck'
+
+async function getSpliceSequences(
+  transcript: AnnotationFeatureSnapshot,
+  getSequence: (start: number, end: number) => Promise<string>,
+): Promise<SpliceSequence[]> {
+  if (!transcript.children) {
+    return []
+  }
+  const exons: AnnotationFeatureSnapshot[] = []
+  for (const [, child] of Object.entries(transcript.children)) {
+    if (child.type === 'exon') {
+      exons.push(child)
+    }
+  }
+  if (exons.length < 2) {
+    return []
+  }
+  exons.sort((a, b) => (a.min < b.min ? -1 : 1))
+
+  const spliceSeq: SpliceSequence[] = []
+  for (let i = 0; i < exons.length - 1; i++) {
+    let fivePrimeMin = exons[i].max
+    let threePrimeMin = exons[i + 1].min
+    if (transcript.strand === -1) {
+      ;[threePrimeMin, fivePrimeMin] = [fivePrimeMin, threePrimeMin]
+      fivePrimeMin -= 2
+    } else {
+      threePrimeMin -= 2
+    }
+
+    let fivePrimeSeq = await getSequence(fivePrimeMin, fivePrimeMin + 2)
+    let threePrimeSeq = await getSequence(threePrimeMin, threePrimeMin + 2)
+    if (transcript.strand === -1) {
+      threePrimeSeq = revcom(threePrimeSeq)
+      fivePrimeSeq = revcom(fivePrimeSeq)
+    }
+
+    spliceSeq.push({ fivePrimeSeq, fivePrimeMin, threePrimeSeq, threePrimeMin })
+  }
+  return spliceSeq
+}
+
+async function checkTranscript(
+  feature: AnnotationFeatureSnapshot,
+  getSequence: (start: number, end: number) => Promise<string>,
+): Promise<CheckResultSnapshot[]> {
+  const checkResults: CheckResultSnapshot[] = []
+
+  const VALID_FIVE_PRIME_SEQ = new Set(['GT'])
+  const VALID_THREE_PRIME_SEQ = new Set(['AG'])
+  const spliceSequences = await getSpliceSequences(feature, getSequence)
+  for (const spliceSequence of spliceSequences) {
+    if (!VALID_FIVE_PRIME_SEQ.has(spliceSequence.fivePrimeSeq.toUpperCase())) {
+      checkResults.push({
+        _id: new ObjectID().toHexString(),
+        name: CHECK_NAME,
+        cause: CAUSES[CAUSES.NonCanonicalSpliceSiteAtFivePrime],
+        ids: [feature._id],
+        refSeq: feature.refSeq.toString(),
+        start: spliceSequence.fivePrimeMin,
+        end: spliceSequence.fivePrimeMin + 2,
+        message: `Unexpected 5′ splice site in "${getPrintableId(feature)}". Expected: ${[...VALID_FIVE_PRIME_SEQ].join('|')}, got: ${spliceSequence.fivePrimeSeq}`,
+      })
+    }
+    if (
+      !VALID_THREE_PRIME_SEQ.has(spliceSequence.threePrimeSeq.toUpperCase())
+    ) {
+      checkResults.push({
+        _id: new ObjectID().toHexString(),
+        name: CHECK_NAME,
+        cause: CAUSES[CAUSES.NonCanonicalSpliceSiteAtThreePrime],
+        ids: [feature._id],
+        refSeq: feature.refSeq.toString(),
+        start: spliceSequence.threePrimeMin,
+        end: spliceSequence.threePrimeMin + 2,
+        message: `Unexpected 3′ splice site in "${getPrintableId(feature)}". Expected: ${[...VALID_THREE_PRIME_SEQ].join('|')}, got: ${spliceSequence.threePrimeSeq}`,
+      })
+    }
+  }
+  return checkResults
+}
+
+function getCauses(): string[] {
+  return Object.values(CAUSES).filter((x) =>
+    Number.isNaN(Number(x)),
+  ) as string[]
+}
+export class TranscriptCheck extends Check {
+  name = 'TranscriptCheck'
+  causes = getCauses()
+  version = 1
+  isDefault = true
+
+  async checkFeature(
+    feature: AnnotationFeatureSnapshot,
+    getSequence: (start: number, end: number) => Promise<string>,
+  ): Promise<CheckResultSnapshot[]> {
+    if (
+      feature.type === 'mRNA' ||
+      feature.type === 'transcript' ||
+      feature.type === 'pseudogenic_transcript'
+    ) {
+      return checkTranscript(feature, getSequence)
+    }
+
+    if (!feature.children) {
+      return []
+    }
+
+    const checkResults: CheckResultSnapshot[] = []
+    for (const child of Object.values(feature.children)) {
+      checkResults.push(...(await this.checkFeature(child, getSequence)))
+    }
+    return checkResults
+  }
+}

package/src/Checks/index.ts

@@ -1 +1,2 @@
 export * from './CDSCheck'
+export * from './TranscriptCheck'

package/src/Checks/util.ts

@@ -0,0 +1,13 @@
+import { type AnnotationFeatureSnapshot } from '@apollo-annotation/mst'
+
+export function getPrintableId(feature: AnnotationFeatureSnapshot): string {
+  const gff_id = feature.attributes?.gff_id?.join(', ')
+  if (gff_id) {
+    return `${gff_id} (_id: ${feature._id.toString()})`
+  }
+  const gff_name = feature.attributes?.gff_name?.join(', ')
+  if (gff_name) {
+    return `${gff_name} (_id: ${feature._id.toString()})`
+  }
+  return `_id: ${feature._id.toString()}`
+}

package/src/GFF3/annotationFeatureToGFF3.ts

@@ -101,7 +101,10 @@ export function annotationFeatureToGFF3(
     score,
     strand: feature.strand ? (feature.strand === 1 ? '+' : '-') : null,
     phase: null,
-    attributes: Object.keys(attributes).length > 0 ? attributes : null,
+    attributes:
+      Object.keys(attributes).length > 0
+        ? (attributes as Record<string, string[]>)
+        : null,
     derived_features: [],
     child_features: prepareChildFeatures(
       feature,

package/src/GFF3/gff3ToAnnotationFeature.test.ts

@@ -3,7 +3,7 @@ import { readFileSync } from 'node:fs'
 import { describe, it } from 'node:test'
 
 import { type AnnotationFeatureSnapshot } from '@apollo-annotation/mst'
-import gff, { type GFF3Feature } from '@gmod/gff'
+import { type GFF3Feature, parseStringSync } from '@gmod/gff'
 import { assert, use } from 'chai'
 import chaiExclude from 'chai-exclude'
 
@@ -116,7 +116,7 @@ function readFeatureFile(fn: string): GFF3Feature[] {
       feature.push(line)
     }
   }
-  const inGff = gff.parseStringSync(feature.join('\n')) as GFF3Feature[]
+  const inGff = parseStringSync(feature.join('\n')) as GFF3Feature[]
   return inGff
 }
 
@@ -127,7 +127,7 @@ export function readAnnotationFeatureSnapshot(
   return JSON.parse(lines) as AnnotationFeatureSnapshot
 }
 
-const [ex1, ex2, ex3, ex4] = readFeatureFile(
+const [ex1, , ex2, , ex3, , ex4] = readFeatureFile(
   'test_data/gene_representations.gff3',
 )
 
@@ -232,7 +232,7 @@ describe('gff3ToAnnotationFeature', () => {
   for (const testCase of testCases) {
     const [description, featureLine, convertedFeature] = testCase
     it(`converts ${description}`, () => {
-      const gff3Feature = gff.parseStringSync(featureLine, {
+      const gff3Feature = parseStringSync(featureLine, {
         parseSequences: false,
       })
       const feature = gff3ToAnnotationFeature(gff3Feature[0])

package/src/GFF3/gff3ToAnnotationFeature.ts

@@ -115,7 +115,7 @@ function convertFeatureAttributes(
   if (attributesCollections.length > 0) {
     for (const attributesCollection of attributesCollections) {
       for (const [key, val] of Object.entries(attributesCollection)) {
-        if (!val || key === 'Parent') {
+        if (key === 'Parent') {
           continue
         }
         const newKey = isGFFReservedAttribute(key) ? gffToInternal[key] : key

package/src/util.ts

@@ -1,5 +1,6 @@
 import { type AnnotationFeature } from '@apollo-annotation/mst'
 import { type Feature } from '@apollo-annotation/schemas'
+import { type IKeyValueMap } from 'mobx'
 
 export function splitStringIntoChunks(
   input: string,
@@ -25,3 +26,39 @@ export function getPrintableId(feature: Feature): string {
   }
   return `_id: ${feature._id.toString()}`
 }
+
+export function attributesToRecords(
+  attributes: IKeyValueMap<readonly string[] | undefined> | undefined,
+): Record<string, string[] | undefined> {
+  const records: Record<string, string[] | undefined> = {}
+  if (!attributes) {
+    return records
+  }
+  for (const [key, value] of Object.entries(attributes)) {
+    records[key] = value?.slice()
+  }
+  return records
+}
+
+export function stringifyAttributes(
+  attributes: Record<string, string[] | undefined> | undefined,
+): string {
+  if (!attributes) {
+    return ''
+  }
+  const str = []
+  for (const [key, value] of Object.entries(attributes)) {
+    let attributeName = key
+    if (attributeName.startsWith('gff_')) {
+      attributeName = attributeName.slice(4)
+      attributeName =
+        attributeName.charAt(0).toUpperCase() + attributeName.slice(1)
+    }
+    if (value) {
+      str.push(`${attributeName}=${value.join(',')}`)
+    } else {
+      str.push(attributeName)
+    }
+  }
+  return encodeURIComponent(str.join(';'))
+}

package/test_data/example01.json

@@ -14,7 +14,7 @@
       "max": 1012,
       "strand": 1,
       "attributes": {
-        "gff_id": ["tfbs10001"]
+        "gff_id": ["tfbs00001"]
       }
     },
     "66e049f17b9cedae9ad890fb": {
@@ -33,7 +33,7 @@
           "max": 1499,
           "strand": 1,
           "attributes": {
-            "gff_id": ["exon10002"]
+            "gff_id": ["exon00002"]
           }
         },
         "66e049f17b9cedae9ad890f7": {
@@ -44,7 +44,7 @@
           "max": 3902,
           "strand": 1,
           "attributes": {
-            "gff_id": ["exon10003"]
+            "gff_id": ["exon00003"]
           }
         },
         "66e049f17b9cedae9ad890f8": {
@@ -55,7 +55,7 @@
           "max": 5500,
           "strand": 1,
           "attributes": {
-            "gff_id": ["exon10004"]
+            "gff_id": ["exon00004"]
           }
         },
         "66e049f17b9cedae9ad890f9": {
@@ -66,7 +66,7 @@
           "max": 9000,
           "strand": 1,
           "attributes": {
-            "gff_id": ["exon10005"]
+            "gff_id": ["exon00005"]
           }
         },
         "66e049f17b9cedae9ad890fa": {
@@ -77,13 +77,13 @@
           "max": 7600,
           "strand": 1,
           "attributes": {
-            "gff_id": ["cds10001"],
+            "gff_id": ["cds00001"],
             "gff_name": ["edenprotein.1"]
           }
         }
       },
       "attributes": {
-        "gff_id": ["mRNA10001"],
+        "gff_id": ["mRNA00001"],
         "gff_name": ["EDEN.1"]
       }
     },
@@ -103,7 +103,7 @@
           "max": 1499,
           "strand": 1,
           "attributes": {
-            "gff_id": ["exon10002"]
+            "gff_id": ["exon00002"]
           }
         },
         "66e049f17b9cedae9ad890fd": {
@@ -114,7 +114,7 @@
           "max": 5500,
           "strand": 1,
           "attributes": {
-            "gff_id": ["exon10004"]
+            "gff_id": ["exon00004"]
           }
         },
         "66e049f17b9cedae9ad890fe": {
@@ -125,7 +125,7 @@
           "max": 9000,
           "strand": 1,
           "attributes": {
-            "gff_id": ["exon10005"]
+            "gff_id": ["exon00005"]
           }
         },
         "66e049f17b9cedae9ad890ff": {
@@ -136,13 +136,13 @@
           "max": 7600,
           "strand": 1,
           "attributes": {
-            "gff_id": ["cds10002"],
+            "gff_id": ["cds00002"],
             "gff_name": ["edenprotein.2"]
           }
         }
       },
       "attributes": {
-        "gff_id": ["mRNA10002"],
+        "gff_id": ["mRNA00002"],
         "gff_name": ["EDEN.2"]
       }
     },
@@ -162,7 +162,7 @@
           "max": 1499,
           "strand": 1,
           "attributes": {
-            "gff_id": ["exon10001"]
+            "gff_id": ["exon00001"]
           }
         },
         "66e049f17b9cedae9ad89102": {
@@ -173,7 +173,7 @@
           "max": 3902,
           "strand": 1,
           "attributes": {
-            "gff_id": ["exon10003"]
+            "gff_id": ["exon00003"]
           }
         },
         "66e049f17b9cedae9ad89103": {
@@ -184,7 +184,7 @@
           "max": 5500,
           "strand": 1,
           "attributes": {
-            "gff_id": ["exon10004"]
+            "gff_id": ["exon00004"]
           }
         },
         "66e049f17b9cedae9ad89104": {
@@ -195,7 +195,7 @@
           "max": 9000,
           "strand": 1,
           "attributes": {
-            "gff_id": ["exon10005"]
+            "gff_id": ["exon00005"]
           }
         },
         "66e049f17b9cedae9ad89105": {
@@ -206,7 +206,7 @@
           "max": 7600,
           "strand": 1,
           "attributes": {
-            "gff_id": ["cds10003"],
+            "gff_id": ["cds00003"],
             "gff_name": ["edenprotein.3"]
           }
         },
@@ -218,19 +218,19 @@
           "max": 7600,
           "strand": 1,
           "attributes": {
-            "gff_id": ["cds10004"],
+            "gff_id": ["cds00004"],
             "gff_name": ["edenprotein.4"]
           }
         }
       },
       "attributes": {
-        "gff_id": ["mRNA10003"],
+        "gff_id": ["mRNA00003"],
         "gff_name": ["EDEN.3"]
       }
     }
   },
   "attributes": {
-    "gff_id": ["gene10001"],
+    "gff_id": ["gene00001"],
     "gff_name": ["EDEN"]
   }
 }