Should HDR regions be considered haplotype-specific?

[PaoloCallipo00]

Hello,

Thank you for developing this great software!

I have a question regarding the interpretation of the SyRI output.
I am analyzing the differences between the two haplotypes of a diploid plant and, in particular, I want to quantify and characterize the hemizygous (haplotype-specific) regions of both haplotypes.

To define these haplotype-specific regions, I currently include NOTAL, INDEL (>100 bp), CPG, CPL, and TDM.
My question is: should HDR also be included?
From my understanding, these regions should also be haplotype-specific since they are not aligned to any region in the opposite haplotype. However, they are not classified as NOTAL because they occur within a SYN block.

Additionally, when inspecting some HDRs in IGV, they sometimes appear to be large INDELs. For example, in this case a 10 kb INDEL is reported as an HDR.
The tracks are HiFi reads from one haplotype realigned to the opposite haplotype.
Is this expected behavior?

Thank you again for providing such a useful tool!

Best regards,
Paolo

[mnshgl0110]

I think, it depends on the project goals. Strictly speaking, a single base-pair insertion is also haplotype-specific sequence. So, with that spirit HDRs are haplotype-specific sequences as well. Syri doesn't report them as NOTALs because the synteny of the neighbouring regions suggests that the region probably had a shared ancestor.