Welcome to “back to sequences” documentation!

back to sequences is a rust program for bioinformaticians. It’s purpose is to find where a set of \(k\)-mers occurs in a set of input genomic sequences.

More precisely:

Given a set \(K\) of \(k\)-mers (fasta / fastq [.gz] format) and a set of sequences (fasta / fastq [.gz] format), this tool will extract the sequences containing some of those kmers.

A minimal (\(m\)) and a maximal (\(M\)) thresholds are proposed. A sequence whose percentage of kmers shared with \(K\) are in \(]m, M]\) is output with its original header + the number of shared kmers + the ratio of shared kmers: ` >original_header 20 6.13 TGGATAAAAAGGCTGACGAAAGGTCTAGCTAAAATTGTCAGGTGCTCTCAGATAAAGCAGTA... ` In this case 20 kmers are shared with the indexed kmers. This represents 6.13% of the kmers in the sequence.

Contents

• Usage
  • Installation
  • back to sequences parameters
• Some basical use cases of back to sequences
  • Basic case
  • Using filters
  • Specifying strands
  • Reading sequences from standard input
  • Using several input read sets
  • Output matching kmers
• Results
  • Benchmark on synthetic data
  • Comparing possible alternatives
  • Benchmark on Tara ocean seawater metagenomic data
• Methods and features
  • Methods
  • Additional features
• Citation