Figure 3.

Processing in STEAK. (A) The software runs as several processes that read different parts of the input SAM or FASTQ file. (B) In the case of a BAM/CRAM file, because the data is read from a single point, a unique process is executed but it spawns several threads, one that accumulates the data in a buffer and the others that process them. TEs are specified in a FASTA file. The trimmed sequence is written in the first FASTQ file. The original sequence of the other read (or the trimmed sequence if there is a match as well) is placed in the second FASTQ file.