search

lh3 / minimap2

A versatile pairwise aligner for genomic and spliced nucleotide sequences
https://lh3.github.io/minimap2
Other
1.8k stars 412 forks source link

Long Inversions not detectable 2.19 to 2.24 #1050

Open DHmeduni opened 1 year ago

DHmeduni commented 1 year ago

I tested 2.17 to 2.24 using an inversion between two large homologous domains. 2.17 and 2.18 detection was possible using structural variant callers. Everything after that was unable to be found. I'll help find the reason if I can, but I don't have the software skills to look under the hood too much. I'm guessing this has to do with the algorithm change in 2.19.

lh3 commented 1 year ago

Do you have the sequence? Thanks

DHmeduni commented 1 year ago

I created a reference sequence with an inversion, then mapped normal gDNA nanopore sequences against it, works with 2.17 and 2.18 using npinv, but not with anything 2.19 or above. If you need the data, I can't supply you with the sequenced data, but with the reference incl. the inversion (several hundred MB) you could use NanoSim to create sequencing data based on the normal reference sequence or inverted reference sequence and then map to the opposite. I'd need a place to upload the inverted reference sequence if you would like. But since we are looking to publish this all, not sure my PI wants to give anything to a third party. I'll have to ask in the meantime. Best Regards, David

On Wed, 26 Apr 2023 at 00:27, Heng Li @.***> wrote:

Do you have the sequence? Thanks

— Reply to this email directly, view it on GitHub https://github.com/lh3/minimap2/issues/1050#issuecomment-1522500372, or unsubscribe https://github.com/notifications/unsubscribe-auth/A7KF4B4DUETNZFTKUIJL5ALXDBFVRANCNFSM6AAAAAAXFJ3NK4 . You are receiving this because you authored the thread.Message ID: @.***>