cmdcolin
commented
4 months ago JBrowse 2 does actually support CIGAR strings. if you generate a PAF file with "minimap2 -c" then it will include the CIGAR string, and jbrowse 2 will render the insertions and deletions
example
there are still challenges though. for example
- we do not detect SNPs in the large CIGAR strings, so those indications, which could be markers to orient oneself, are not available. I have considered using the #3378 as a way to help
- if there is a large e.g. 10_000 base match, it can be difficult to see where you are in that large match. in this case, adding just 'accessory grid functions' that don't even indicate sequence features, just sort of aligned blocks, can help. similar request https://github.com/GMOD/jbrowse-components/issues/3002
lybCNU
Is your feature request related to a problem? Please describe. I've encountered a problem when using JBrowse 2's linear synteny view to observe synteny results generated by minimap2. The issue arises when dealing with long syntenic regions, as it becomes challenging to maintain detail and orientation when focusing on specific genes. This makes it difficult to accurately align and compare specific regions across different genomes, especially when looking for minor structural variations or similarities between them.
Describe the solution you'd like I would like JBrowse 2 to introduce a feature similar to the "grid lines for aligned regions" found in the traditional GBrowse_syn for its linear synteny view. Specifically, I wish for the linear synteny view to be able to read the CIGAR information from PAF files generated by minimap2 and use this to generate grid lines for aligned regions. This enhancement would allow users to clearly see the boundaries of each aligned region, facilitating easy focus on specific genes or areas even within extensive syntenic regions.
Describe alternatives you've considered Aside from directly supporting CIGAR information in the linear synteny view, I've also considered the possibility of preprocessing the PAF file with external tools to mark the boundaries of aligned regions before importing it into JBrowse. However, this approach is indirect and adds complexity for the user. Implementing this feature directly within JBrowse would be more convenient and intuitive for users.
Additional context Minimap2 is a widely used sequence alignment tool, and its output PAF files contain valuable alignment information, including CIGAR strings. The CIGAR information is crucial for understanding the alignment relationships between sequences, helping researchers accurately identify insertions, deletions, and substitutions between sequences. To facilitate the parsing of CIGAR strings within JBrowse 2, I suggest considering the use of the Bio::Cigar Perl module, which is designed to assist in CIGAR string analysis. This could potentially streamline the development of the proposed feature by leveraging existing libraries designed for sequence analysis. Enhancing JBrowse 2's linear synteny view with this capability would significantly increase its utility for genomic alignment analysis.