New version of LR-splitpipe:
demultiplex.py
- Improves performance on larger sets of reads by reading and processing reads in chunks
- Option to just run the linker scoring so that linker mismatch can be chosen at later steps (
score_linkersmode) - Allows for processing of multiple fastq files to be aggregated at the barcode correction step (subsequent usage of
find_bcsandprocess_bcsstep) - Added options to tune number of allowable mismatches in each linker
- Improved QC
- Removed option to filter on corresponding Illumina data
- Removed option to filter based on read count per barcpde
- Added UMI deduplication with keep the longest read heuristic
add_bam_tag.py
- Added option to append a suffix to each barcode. Useful if combining multiple SAM files from different experimentss down the line that might have overlapping barcodes