search

schatzlab / genomescope

Fast genome analysis from unassembled short reads
Apache License 2.0
258 stars 57 forks source link

High heterozygous? #118

Open shehongbing opened 10 months ago

shehongbing commented 10 months ago

Hi, i estimate the genome using HiFi reads and the obtain high heterozygous (7%). However, i can not infer the two obviously peaks.

linear_plot linear_plot

mschatz commented 10 months ago

I agree the fit is questionable. Do you have any information about the species or the assembly? Can you send the link to the genomescope webpage (or send the histo file)

Thanks! Mike

On Sun, Jan 14, 2024 at 6:16 AM shehongbing @.***> wrote:

Hi, i estimate the genome using HiFi reads and the obtain high heterozygous (7%). However, i can not infer the two obviously peaks.

linear_plot.png (view on web) https://github.com/schatzlab/genomescope/assets/49432965/58124d9e-20aa-4509-9deb-3bf1dd13ede7 linear_plot.png (view on web) https://github.com/schatzlab/genomescope/assets/49432965/b1d3e258-dae1-4a77-be87-1d7f7057ca98

— Reply to this email directly, view it on GitHub https://github.com/schatzlab/genomescope/issues/118, or unsubscribe https://github.com/notifications/unsubscribe-auth/AABP345FIOKUY3Y3HZCG6ELYOPSFRAVCNFSM6AAAAABB2EHX5SVHI2DSMVQWIX3LMV43ASLTON2WKOZSGA4DANZUGY3TMOA . You are receiving this because you are subscribed to this thread.Message ID: @.***>

shehongbing commented 10 months ago

Hi, Mike, Thanks for your response. The genome of the species has not been reported. We used flow cytometry to assess the genome size of 930 Mb. I attached the link to the genomescope webpage below.

http://qb.cshl.edu/genomescope/genomescope2.0/analysis.php?code=srATRMT8GOQqWNd6S9zq

Thanks Hongbing

mschatz commented 8 months ago

Hi Hongbing

I agree there was a bad fit with the model. I can get a slightly better fit by setting "Average k-mer coverage for polyploid genome" to 9. This gives the model fitting algorithm some help in determining which peak is which. With this parameter, the heterozygosity drops to almost nothing and the genome size increases to 1.4Gb. But this fit still isnt as good as I would like and there is still a fairly large gap between the model and the observed counts. I think it would really help to use deeper coverage. Is this at all possible?

Here are the updated results: http://qb.cshl.edu/genomescope/genomescope2.0/analysis.php?code=GL5JdZG5ro7wKPbGfxGj

Good luck! Mike

On Sun, Jan 14, 2024 at 8:01 PM shehongbing @.***> wrote:

Hi, Mike, Thanks for your response. The genome of the species has not been reported. We used flow cytometry to assess the genome size of 930 Mb. I attached the link to the genomescope webpage below.

http://qb.cshl.edu/genomescope/genomescope2.0/analysis.php?code=srATRMT8GOQqWNd6S9zq

Thanks Hongbing

— Reply to this email directly, view it on GitHub https://github.com/schatzlab/genomescope/issues/118#issuecomment-1891147749, or unsubscribe https://github.com/notifications/unsubscribe-auth/AABP34ZFKIVKEGD4G3Y7WQLYOR5XXAVCNFSM6AAAAABB2EHX5SVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQOJRGE2DONZUHE . You are receiving this because you commented.Message ID: @.***>