Home → CLC software: Important notifications → Issues affecting only older versions of products → Some PacBio reads and other reads longer than 500bp are included in mappings despite not passing length and similarity cut-offs
2.28. Some PacBio reads and other reads longer than 500bp are included in mappings despite not passing length and similarity cut-offs
An issue with read mappings was found where length and similarity fraction cut-offs were ignored for the small subset of PacBio reads and reads of other types longer than 500bp where:
- the best alignment of the read involved a stretch of residues that exactly match the reference sequence, and where
- no extension from the alignment seed was possible.
When reads are affected by this problem, they are included in read mappings when they should not be. Such reads will have only a short section of the total read that matches the reference. When the mapping is viewed, such reads are recognized by their long unaligned ends.
For variant detection, we expect little or no effect on outcome. This issue is expected to affect only a small subset of reads that are over 500bp, and for those reads, only a small stretch of matches are added to the mapping. No mismatches are added. Most of an affected mapped read will consist of unaligned ends, and such areas do not contribute to variant calling.
For those mapping contigs to a reference, there is the possibility of some contigs mappings very badly. If affected this way, it is likely to be immediately obvious when the mapping is viewed. A badly mapped contig would have only a short length of residues matching the reference, with most of the contig's residues being unaligned.
- CLC Genomics Workbench 10.0 through 10.1.1
- Biomedical Genomics Workbench 4.0 through 4.1.1
- CLC Genomics Server 9.0 through 9.1.1
- CLC Assembly Cell 5.0, 5.0.1, 5.0.2, and 5.0.3