szpiech / selscan

Haplotype based scans for selection
GNU General Public License v3.0
115 stars 33 forks source link

Detecting natural selection signals in selfing populations #100

Open Axolotl233 opened 1 year ago

Axolotl233 commented 1 year ago

Hi community,

I want to know if selscan can detect natural selection signals in inbreeding populations using whole genome SNP data, despite the long haplotype distances caused by inbreeding and a significant hitchhiking effect, which may weaken the power of detection.

Is this feasible if I use the cross approaches (like xp-nsl or xp-ehh) to detect selection signals between the two populations?

thank you!

szpiech commented 1 year ago

Hmm, this is an interesting question. I haven't tested these stats in any of these types of populations empirically nor through simulations. I did try my other method, saltiLASSI, on arabidopsis data, and it seemed to give sensible results (caveat, I know essentially nothing about arabidopsis biology). Since that method is also based on haplotypes, I might guess guess that these could work too. However, I would first want to ensure that the entire population is engaging in the same rates of selfing, on average, and honestly I'd probably want to do some simulations to confirm it works well before going too far down the analysis hole.

Zachary

On Wed, Jul 26, 2023 at 12:51 PM Axolotl233 @.***> wrote:

Hi community,

I want to know if selscan can detect natural selection signals in inbreeding populations using whole genome SNP data, despite the long haplotype distances caused by inbreeding and a significant hitchhiking effect, which may weaken the power of detection.

Is this feasible if I use the cross approaches (like xp-nsl or xp-ehh) to detect selection signals between the two populations?

thank you!

— Reply to this email directly, view it on GitHub https://github.com/szpiech/selscan/issues/100, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABAKRQVGUSVQ2BGZKRIF5N3XSFDINANCNFSM6AAAAAA2Y32QVI . You are receiving this because you are subscribed to this thread.Message ID: @.***>