Reporting and filtering

The resulting alignments are output to a single CAS file4.3 using the -o/--output parameter:

clc_mapper -d reference.fa -q reads.fastq -o result.cas

By default, if there are multiple, equally good alignments for a particular read, the number of alignments is stored, but detailed alignment information is only stored explicitly for one of them. Should you wish to change this behavior, you can use the following pair of parameters to do so:

The -t/--maxalign parameter determines the maximum number of alignments that are to be stored explicitly. By default, only one is stored explicitly.

The -r/--repeat parameter accepts one of two arguments: random or ignore. By default, it is set to random, which means that the explicitly stored alignments are sampled, randomly, from the total set of alignments found for a given read. The ignore argument results in reads with more alignments than can be explicitly stored are ignored and reported as unmapped.

As an example, suppose that you only want to report unique hits as mapped, i.e. ignore alignments with more than one hit. Then you can use the following combination:

clc_mapper ... -t 1 -r ignore ...



Footnotes

... file4.3
The CAS file is a compact, binary file containing information about the alignment(s) of the individual reads. The read and reference sequences are not stored explicitly, but rather the paths to the input files are stored.