Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

feat: allow TrimFastq to specify a length per input FASTQ #928

Merged
merged 3 commits into from
Aug 16, 2023
Merged

feat: allow TrimFastq to specify a length per input FASTQ #928

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
afa97e6
feat: allow TrimFastq to specify a length per input FASTQ
nh13 Aug 16, 2023
279dfcf
Apply suggestions from code review
nh13 Aug 16, 2023
a789098
review fixups
nh13 Aug 16, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
11 changes: 7 additions & 4 deletions src/main/scala/com/fulcrumgenomics/fastq/TrimFastq.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -46,26 +46,29 @@ import com.fulcrumgenomics.sopt._
class TrimFastq
( @arg(flag='i', doc="One or more input fastq files.") val input: Seq[PathToFastq],
@arg(flag='o', doc="A matching number of output fastq files.") val output: Seq[PathToFastq],
@arg(flag='l', doc="Length to trim reads to.") val length: Int,
@arg(flag='l', doc="Length to trim reads to (either one per input fastq file, or one for all).") val length: Seq[Int],
@arg(flag='x', doc="Exclude reads below the trim length.") val exclude: Boolean = false
) extends FgBioTool with LazyLogging {

validate(input.size == output.size, "Number of input and output files must match.")
validate(length.size == 1 || input.size == length.size, "Number of lengths must be one or match the number of input files.")

override def execute(): Unit = {
var discarded: Long = 0
val progress = new ProgressLogger(this.logger, noun="records", verb="Wrote")

val lengths = if (this.length.size == 1) List.fill(this.input.size)(this.length.head) else this.length

nh13 marked this conversation as resolved.
Show resolved Hide resolved
val sources = input.map(FastqSource(_))
val writers = output.map(FastqWriter(_))
while (allHaveNext(sources)) {
val recs = sources.map(_.next())
if (exclude && recs.exists(_.length < length)) {
if (exclude && recs.zip(lengths).exists { case (rec, length) => rec.length < length }) {
discarded += 1
}
else {
writers.iterator.zip(recs.iterator).foreach { case(w, r) =>
w.write(r.trimmedTo(length))
writers.lazyZip(recs).lazyZip(lengths).foreach { case (w: FastqWriter, r: FastqRecord, l: Int) =>
w.write(r.trimmedTo(l))
progress.record()
}
}
Expand Down
34 changes: 25 additions & 9 deletions src/test/scala/com/fulcrumgenomics/fastq/TrimFastqTest.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -59,9 +59,9 @@ class TrimFastqTest extends UnitSpec {
}

"TrimFastq" should "trim a single file and not discard any records" in {
val (r1, r2) = fqFiles
val (r1, _) = fqFiles
val out = makeTempFile("trimmed.", ".fq")
new TrimFastq(input=Seq(r1), output=Seq(out), length=15, exclude=false).execute()
new TrimFastq(input=Seq(r1), output=Seq(out), length=Seq(15), exclude=false).execute()
val r1Map = FastqSource(out).map(r => r.name -> r).toMap
r1Map.size shouldBe 3
r1Map("10x10").length shouldBe 10
Expand All @@ -70,18 +70,18 @@ class TrimFastqTest extends UnitSpec {
}

it should "trim a single file and discard 2 records" in {
val (r1, r2) = fqFiles
val (r1, _) = fqFiles
val out = makeTempFile("trimmed.", ".fq")
new TrimFastq(input=Seq(r1), output=Seq(out), length=15, exclude=true).execute()
new TrimFastq(input=Seq(r1), output=Seq(out), length=Seq(15), exclude=true).execute()
val r1Map = FastqSource(out).map(r => r.name -> r).toMap
r1Map.size shouldBe 1
r1Map("20x20").length shouldBe 15
}

it should "trim a single file and discard 0 records because they are all long enough" in {
val (r1, r2) = fqFiles
val (r1, _) = fqFiles
val out = makeTempFile("trimmed.", ".fq")
new TrimFastq(input=Seq(r1), output=Seq(out), length=5, exclude=true).execute()
new TrimFastq(input=Seq(r1), output=Seq(out), length=Seq(5), exclude=true).execute()
val r1Map = FastqSource(out).map(r => r.name -> r).toMap
r1Map.size shouldBe 3
r1Map("10x10").length shouldBe 5
Expand All @@ -92,7 +92,7 @@ class TrimFastqTest extends UnitSpec {
it should "not trim or discard any reads" in {
val (r1, r2) = fqFiles
val (r1Out, r2Out) = (makeTempFile("r1out.", ".fq"), makeTempFile("r2out.", ".fq"))
new TrimFastq(input=Seq(r1, r2), output=Seq(r1Out, r2Out), length=25, exclude=false).execute()
new TrimFastq(input=Seq(r1, r2), output=Seq(r1Out, r2Out), length=Seq(25), exclude=false).execute()
val r1Map = FastqSource(r1Out).map(r => r.name -> r).toMap
val r2Map = FastqSource(r2Out).map(r => r.name -> r).toMap
r1Map.size shouldBe 3
Expand All @@ -108,7 +108,7 @@ class TrimFastqTest extends UnitSpec {
it should "trim but not discard some reads" in {
val (r1, r2) = fqFiles
val (r1Out, r2Out) = (makeTempFile("r1out.", ".fq"), makeTempFile("r2out.", ".fq"))
new TrimFastq(input=Seq(r1, r2), output=Seq(r1Out, r2Out), length=15, exclude=false).execute()
new TrimFastq(input=Seq(r1, r2), output=Seq(r1Out, r2Out), length=Seq(15), exclude=false).execute()
val r1Map = FastqSource(r1Out).map(r => r.name -> r).toMap
val r2Map = FastqSource(r2Out).map(r => r.name -> r).toMap
r1Map.size shouldBe 3
Expand All @@ -124,12 +124,28 @@ class TrimFastqTest extends UnitSpec {
it should "trim some reads and discard others by pair in" in {
val (r1, r2) = fqFiles
val (r1Out, r2Out) = (makeTempFile("r1out.", ".fq"), makeTempFile("r2out.", ".fq"))
new TrimFastq(input=Seq(r1, r2), output=Seq(r1Out, r2Out), length=15, exclude=true).execute()
new TrimFastq(input=Seq(r1, r2), output=Seq(r1Out, r2Out), length=Seq(15), exclude=true).execute()
val r1Map = FastqSource(r1Out).map(r => r.name -> r).toMap
val r2Map = FastqSource(r2Out).map(r => r.name -> r).toMap
r1Map.size shouldBe 1
r2Map.size shouldBe r1Map.size
r1Map("20x20").length shouldBe 15
r2Map("20x20").length shouldBe 15
}

it should "trim some reads and discard others with FASTQ specific lengths" in {
nh13 marked this conversation as resolved.
Show resolved Hide resolved
val (r1, r2) = fqFiles
val (r1Out, r2Out) = (makeTempFile("r1out.", ".fq"), makeTempFile("r2out.", ".fq"))
new TrimFastq(input = Seq(r1, r2), output=Seq(r1Out, r2Out), length = Seq(10, 15), exclude = false).execute()
val r1Map = FastqSource(r1Out).map(r => r.name -> r).toMap
val r2Map = FastqSource(r2Out).map(r => r.name -> r).toMap
r1Map.size shouldBe 3
r2Map.size shouldBe r1Map.size
r1Map("10x10").length shouldBe 10
r1Map("10x20").length shouldBe 10
r1Map("20x20").length shouldBe 10
r2Map("10x10").length shouldBe 10
r2Map("10x20").length shouldBe 15
r2Map("20x20").length shouldBe 15
}
}