This mode is used to merge several compact alignment files to an alignment file. Merge is used when assembling results after searching by chromosome or by transcripts. It is especially useful if the reference does not fit into memory of a computer used for alignment, for example, when searching against transcripts. Each input file must have been generated by searching the same set of reads against different subset of reference sequences. Merging consists of putting back the results as if the set of reads had been searched against the combined reference. Several merging strategies are supported by this class. For instance, one strategy assumes that each set of reads provided as input was searched against a different chromosome (or contig). Another strategy assumes that the reads were aligned to cDNA/transcript individual reference sequences. It is implemented by edu.cornell.med.icb.goby.modes.TODOMode.java.

Mode Parameters

The following options are available in this mode

FlagArgumentsRequiredDescription
(-o|--output)outputyesThe output filenname to write the merged alignments to.
(-k|--k)knoThe k values for the merge. If gene-transcript-map-file is defined, the alignment was against transcripts. k defines the maximum number of genes that a read may match. Best value is probably 1. If gene-transcript-map-file is NOT defined, the alignment was against a reference genome. k defines the maximum number of duplicates allowed for the top quality score when there are duplicates. Best value is probably 2.
(-g|--gene-transcript-map-file)gene-transcript-map-filenoA TSV file that specifies the gene-transcript mapping.
n/ainputyesThe input alignment files to merge.