Add splitkmers command

Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "umgap"
-version = "0.3.3"
+version = "0.3.4"
 authors = ["Felix Van der Jeugt <[email protected]>",
            "Stijn Seghers <[email protected]>",
            "Niels De Graef <[email protected]>",

src/args.rs

@@ -187,6 +187,10 @@ pub enum Opt {
 	#[structopt(name = "printindex")]
 	PrintIndex(PrintIndex),
 
+	/// Splits each protein sequence in a FASTA format into a list of kmers.
+	#[structopt(name = "splitkmers")]
+	SplitKmers(SplitKmers),
+
 	/// Write an FST index of stdin on stdout.
 	#[structopt(name = "buildindex")]
 	BuildIndex,
@@ -550,6 +554,14 @@ pub struct PrintIndex {
 	pub fst_file: PathBuf,
 }
 
+/// Splits each taxon id + protein sequence pair in a CSV format into a list of kmers.
+#[derive(Debug, StructOpt)]
+pub struct SplitKmers {
+	/// The K in K-mers
+	#[structopt(short = "k", long = "length", default_value = "9")]
+	pub length: usize,
+}
+
 error_chain! {
 	errors {
 		/// Unparseable Frame

src/dna/translation.rs

@@ -225,5 +225,4 @@ mod tests {
 	fn test_number_of_codons() {
 		assert_eq!(64, CodonIterator::new().count());
 	}
-
 }

src/main.rs

@@ -71,6 +71,7 @@ quick_main!(|| -> Result<()> {
 		args::Opt::Report(args)          => report(args),
 		args::Opt::BestOf(args)          => bestof(args),
 		args::Opt::PrintIndex(args)      => printindex(args),
+		args::Opt::SplitKmers(args)      => splitkmers(args),
 		args::Opt::BuildIndex            => buildindex(),
 		args::Opt::CountRecords          => countrecords(),
 	}
@@ -715,6 +716,27 @@ fn printindex(args: args::PrintIndex) -> Result<()> {
 	Ok(())
 }
 
+fn splitkmers(args: args::SplitKmers) -> Result<()> {
+	let mut reader = csv::ReaderBuilder::new()
+	                                    .has_headers(false)
+	                                    .delimiter(b'\t')
+	                                    .from_reader(io::stdin());
+
+	let mut writer = csv::WriterBuilder::new()
+	                                    .delimiter(b'\t')
+	                                    .from_writer(io::stdout());
+
+	for record in reader.deserialize() {
+		let (tid, sequence): (TaxonId, String) = record?;
+		if sequence.len() < args.length { continue }
+		for kmer in sequence.as_bytes().windows(args.length) {
+			writer.serialize((String::from_utf8_lossy(kmer), tid))?;
+		}
+	}
+
+	Ok(())
+}
+
 fn buildindex() -> Result<()> {
 	let mut reader = csv::ReaderBuilder::new()
 	                                    .has_headers(false)

src/rmq/rmq.rs

@@ -52,7 +52,7 @@ impl<T: Ord + Display> RMQ<T> {
 			      .min_by_key(|&(_, val)| val)
 			      .expect("So, it has come to this.")
 			      .0 + i * size()
-			    })
+		     })
 		     .collect()
 	}
 

src/tree/tree.rs

@@ -56,7 +56,7 @@ impl<T: Default + Copy> Tree<T> {
 			                         set.iter()
 			                            .map(|&tid| Tree::create(tid, children, taxons))
 			                            .collect()
-			                        })
+		                         })
 		                         .unwrap_or(Vec::new()) }
 	}