search

polyactis / Accucopy

Accucopy is a computational method that infers Allele-Specific Copy Number alterations from low-coverage low-purity tumor sequencing data.
https://www.yfish.org/software/Accucopy
GNU General Public License v3.0
15 stars 4 forks source link

Changing window size #1

Closed NicoleGruenheit closed 3 years ago

NicoleGruenheit commented 4 years ago

Hi,

I tested the accucopy docker with a dataset of 100 samples, but no breakpoints were found for 80 of those. So I tried to use a different window size but it seems that somewhere the window size is hard coded as it tries to find the file 'chr22.ratio.w500.csv.gz' in the next step no matter what the actual window size is. Of course, this file does not exist as for a window size of 250 it's actually 'chr22.ratio.w250.csv.gz' and the computation fails. Is it possible to fix this? Is there anything else i could try to get a Tumor purity value for those 80 samples?

Cheers, Nicole

polyactis commented 4 years ago

Hi Nicole,

We will add a test case for window size 250 in our internal testing and fix relevant bugs.

Can you give us an example of the output for one of your 80 samples? All files in "infer_result_debug", except "het_snp.tsv.gz" and "model_selection_log/" folder, which are too big. These files do not contain any patient/human privacy data. You can email them to my personal email polyactis@gmail.com if you have any concern.

Yu

On Tue, Aug 4, 2020, 8:05 PM NicoleGruenheit notifications@github.com wrote:

Hi,

I tested the accucopy docker with a dataset of 100 samples, but no breakpoints were found for 80 of those. So I tried to use a different window size but it seems that somewhere the window size is hard coded as it tries to find the file 'chr22.ratio.w500.csv.gz' in the next step no matter what the actual window size is. Of course, this file does not exist as for a window size of 250 it's actually 'chr22.ratio.w250.csv.gz' and the computation fails. Is it possible to fix this? Is there anything else i could try to get a Tumor purity value for those 80 samples?

Cheers, Nicole

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/polyactis/Accucopy/issues/1, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAF7C2PFHXAABYNOPYWP6Z3R672RTANCNFSM4PUKO3YQ .

polyactis commented 4 years ago

BTW, 'chr22.ratio.w500.csv.gz' is a minor bug. During debug mode (--debug or -d), Accucopy will output a coverage plot between tumor and normal after normalization, only for chr22 (no other chromosomes). The commandline has 500 embedded in it. Other programs are not affected. If you remove -d, it should go through. We'll fix this bug in the next release.

On Wed, Aug 5, 2020 at 9:00 AM Yu Huang polyactis@gmail.com wrote:

Hi Nicole,

We will add a test case for window size 250 in our internal testing and fix relevant bugs.

Can you give us an example of the output for one of your 80 samples? All files in "infer_result_debug", except "het_snp.tsv.gz" and "model_selection_log/" folder, which are too big. These files do not contain any patient/human privacy data. You can email them to my personal email polyactis@gmail.com if you have any concern.

Yu

On Tue, Aug 4, 2020, 8:05 PM NicoleGruenheit notifications@github.com wrote:

Hi,

I tested the accucopy docker with a dataset of 100 samples, but no breakpoints were found for 80 of those. So I tried to use a different window size but it seems that somewhere the window size is hard coded as it tries to find the file 'chr22.ratio.w500.csv.gz' in the next step no matter what the actual window size is. Of course, this file does not exist as for a window size of 250 it's actually 'chr22.ratio.w250.csv.gz' and the computation fails. Is it possible to fix this? Is there anything else i could try to get a Tumor purity value for those 80 samples?

Cheers, Nicole

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/polyactis/Accucopy/issues/1, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAF7C2PFHXAABYNOPYWP6Z3R672RTANCNFSM4PUKO3YQ .

polyactis commented 4 years ago

Hi Nicole,

One issue we noticed with other users is that their bam files were not aligned with our reference genome. Mostly this is fine as long as it's the same version. However, when the order of chromosomes in the bam header (i.e. chrM is the first entry instead of chr1 in one user's case) differs from the order in our files (.dict), then problems arise. You can give the bam a new header.

We are working to see if we can resolve it in the code.

polyactis commented 4 years ago

We have released a new docker version of Accucopy that solved the different-order-of-chromosomes bug. You can try it https://hub.docker.com/r/polyactis/accucopy .

But if your bam files use a ref genome that is quite different from the ones provided by us. I recommend you to create a new reference genome folder, https://www.yfish.org/display/PUB/Accucopy#Accucopy-3.7Makeyourownreferencegenomepackage.

NicoleGruenheit commented 4 years ago

Hi Yu,

thanks for the incredibly fast reply. Just using the new docker didn't help but I'll try your other suggestions and will come back to you ASAP.

Cheers, Nicole

polyactis commented 3 years ago

Patched. Accucopy can now handle different window sizes.