Seq Crumbs aims to be a collection of small sequence processing utilities.

Seq Crumbs is modeled after the Unix command line text processing utilities so every utility tries to perform a specific task and most of them take a sequence file as input and create a new processed sequence file as output. This design encorages the assembly of the Seq Crumbs utilities with Unix pipes.

You can find more information about seq_crumbs in the seq_crumbs web site.

sff_extract

It extracts the reads from a SFF file used by 454 and Ion Torrent.

split_matepairs

It splits the mate-pairs separated by an oligo sequence.

filter_by_quality

It filters the sequences according to its mean quality.

filter_by_length

It filters sequences according to a length threshold.

filter_by_name

It filters the sequences with a list of names given in a file.

filter_by_blast

It filters the sequences using blast

filter_by_bowtie2

It filters the sequences using bowtie2

filter_by_complexity

It filters the sequences according to its complexity.

trim_by_case

It trims the sequences according to its case.

trim_edges

It removes a fixed number of residues from the sequence edges.

trim_quality

It removes with a sliding windows regions of low quality in the edges

trim_blast_short

It removes oligonucleotides by using the blast-short algoritm

convert_format

It changes between the different supported sequence formats.

guess_seq_format

It guesses the format of a file, including Sanger and Illumina fastq formats.

cat_seqs

It concatenates one or several input sequence files, that can be in different formats, into one output.

seq_head

It outputs only the first sequences of the given input.

sample_seqs

It does a random sampling of the input sequences.

change_case

It modifies the letter case of the sequences. The case can be converted to lower or upper case or swapped.

pair_matcher

It filters out orphaned read pairs.

interleave_pairs

It interleaves two ordered paired read files.

deinterleave_pairs

It splits a ordered file of paired reads into two files, one for every end.

calculate_stats

It generates basic statistics for the given sequence files

count_seqs

It counts the number of sequences and the total length for the given files

orientate_transcripts

It reverse complements transcripts according to polyA, ORF or blast hits

All seq crumbs try to share a consistent interface. Most Seq Crumbs take can their input from the standard input to be able to work with Unix Pipes. Alternatively several input sequence files can be provided as a list of arguments. By default they throw their output to the standard output, although this behaviour can be changed with the -o parameter (or --outfile).

seq_crumbs supports compressed gzip, BGZF and bzip2 files. When used as input it autodetects the compressed files. It can also generate compressed outputs.

The sequence formats accepted by seq_crumbs are the ones supported by Biopython's SeqIO module. As output only Sanger and Illumina fastq and fasta files are supported.

seq_crumbs can take advantage of the multiprocessor computers by splitting the computational load into several processes.

The filtering seq crumbs can be aware of the paired reads and can filter both reads of the pairs together.

Seq Crumbs depends on Python 2.7 and Biopython. The installation manual is located in the INSTALL document.

seq_crumbs relies heavily on Biopython and without this free software project it won't be able to provide some of its functionalities.

Biopieces is a project with a scope similar to seq_crumbs. Biopieces is a great software project capable of working with different kinds of biological data using Unix Pipes. seq_crumbs tries to be more limited in its scope limiting itself only to sequence files and thus providing a somewhat simpler interface.

Another software very similar in the approach to seq_crumbs is the nice fastx collection.

Other related software: PRINSEQ, ea-utils, Pyrocleaner, Sequence Cleaner, lucy, NGS QC Toolkit, scythe, sickle, cutadapt, trimomatic and FastQC.

Seq Crumbs is free software. Licensed mainly under the General Public License (GPL). For more details on the licensing take a look at the LICENSE.txt file.