FEEDBACK WANTED: Installation and running instructions for beta-version of Smudgeplot 0.3.0 Oriel

[KamilSJaron]

This issue is for the brave beta-testers that would like to test the beta-version of the lastest version of Smudgeplot 0.3.0 Oriel.

Fist, get the smudgeplot repository

git clone https://github.com/KamilSJaron/smudgeplot.git

then pull also the development branch

cd smudgeplot
git branch sploidyplot #create a branch for the developmental version
git checkout sploidyplot #checkout to that branch
git pull origin sploidyplot #pull the new version locally

Now you downloaded the beta-version. There is a readme file with installation instructions and everything you need to know to run the beta-version (hopefully, it's beta after all).

Let me know if it works for you!

[sujaikumar]

or in one command :-)

git clone -b sploidyplot https://github.com/KamilSJaron/smudgeplot

[KamilSJaron]

I should have thought there is a simpler solution, thanks Sujai!

[bm-silva]

Hi @KamilSJaron,

I successfully installed and ran smudgeplot 0.3.0 without any issues. However, I have some inquiries regarding the FastK and PloidyPlot parameters:

FastK:
-t: Generates a table of sorted k-mers and counts greater than or equal to the specified level.

PloidyPlot:
-e: Sets the count threshold below which k-mers are considered erroneous.

smudgeplot.py:
-n N: Sets the expected haploid coverage (defaults to estimation from data).

If I misuse these arguments, could it lead to erroneous ploidy predictions? Regarding the -n flag, can it be left blank? Additionally, could a highly repetitive genome contribute to inaccurate predictions?

Thank you for your assistance!

[KamilSJaron]

Hi @bm-silva, 100% yes.

-t can be picked fairly small (4 is fairly common); it's good to see the low coverage kmers in at least the coverage histogram
-e is the parameter L in smudgeplot.py hetmers interface. That should be decided upon inspection of the histogram; usually >10 for any dataset that was intended for reference genome sequencing (if you will get a messy smudgeplot it probably means you would not reconstruct the genome well anyway)
-n is a parameter I would not recommend to use unless you think you really know what the coverage should be.

[bm-silva]

Thank you for the clarification and recommendations.

Regarding the -n parameter, would it be feasible to utilize the kmercov value from the genomescope2 model as the input for this flag?

[KamilSJaron]

Of course you can, the question is if you trust it. There is no harm in doing both. The smudgeplot rule of thumb is "if it looks right, it probably is"