question regarding ascat parameters

[apsteinberg]

Hello,

Thanks for building this great pipeline -- it looks very useful. I was wondering why for your ASCAT module you're setting the bin size to 2000? The reason I ask is I see in the ascat docs here that they recommend settings to 500. I guess some of these parameters are dependent on the type of long-read sequencing that was performed on the samples?

Thanks,
Asher

[ljwharbers]

Hi Asher,

Sorry for the late reply and thanks a lot for the kind words!

Currently, we have the defaults for ASCAT set to what we have seen works best in our (mostly pacbio but also some ONT) samples. The 'bin size' mostly depends on how long your reads are. Longer reads would require a higher bin size.
However, these defaults definitely do not work well for every sample and we have not done very in-depth testing of what the absolute best bin size would be. We therefore definitely encourage you to see what works best for you.

Don't hesitate to reach out with any further questions, we are still working on getting our first 1.0.0 release with proper documentation.

Cheers,
Luuk

[apsteinberg]

Hi Luuk,

No worries at all! Very cool, we have mostly ONT data so I'll play around with the bin size as you have suggested to see what works.

Sounds great, looking forward to the 1.0.0 release!

Cheers,
Asher

[ljwharbers]

Thanks a lot! Also feel free to let us know if you have a specific bin size (or other settings) that work best for your data and we can have a look to change the defaults.

Closing this for now but feel free to reopen!

[ljwharbers]

@apsteinberg Just tagging you because we changed some ASCAT defaults. We're not completely sure why but initially we figured that, since there is no PCR amplification of our DNA, the GC correction is not required. However, we now noticed GC correction vastly improves the quality of the ASCAT copy number profiles. We are currently merging the PR so feel free to try it out on the latest dev branch.

nextflow run IntGenomicsLab/lrsomatic -r dev ...

don't forget to pull the latest if you have run the dev branch before using:

nextflow pull IntGenomicsLab/lrsomatic -r dev

Thanks again for using the pipeline and let us know if there are any improvements you would like to see!

[apsteinberg]

Hi @ljwharbers, my apologies that I never responded here! I am wondering if these changes were merged in the end or should I run on dev? I'd like to test this out on some of our samples.

Thanks,
Asher

[ljwharbers]

I would definitely run it on dev. We made a decent amount of changes (especially improvement for tumour only samples).
Do keep in mind that we haven't tested all these changes yet on full sized samples.

If you want to play it save, you can run the commit where we merged the ASCAT changes, which is 9eda20b. You should be able to run nextflow run intgenomicslab/lrsomatic -r 9eda20b.
If you do use the dev branch, please let me know if everything goes well in that case!

[apsteinberg]

Excellent, sounds good thanks for the quick response!

Asher