Deeptools questions

94 views
Skip to first unread message

Iddir Yasmine

unread,
Dec 10, 2019, 12:31:19 PM12/10/19
to deep...@googlegroups.com

Dear Deeptools developper, 


Thanks for having developed this amazing tool. Sorry, If you have already replied for this questions, but I want to know how does bamCoverage tool and computeMatrix reference-point TSS works  exactly please?

So far in my understoond, bamCoverage does normalization step, but how? computeMatrix doesn't normalize rigth?  

I want to know please, how plotHeatmap works and what does the score refer to? and also how does plotProfile works and what are the scores and why are they different from those of heatmap? 

Sorry in advanced if mentionned before.


Best regards,


Yasmine Iddir
Bioinformaticienne

Équipe SiRIC Recherche Translationnelle en Oncologie Pédiatrique U830

Institut Curie

26 rue d'Ulm

Bureau 1B-03 
75248 Paris Cedex 05
+33 (0)1 56 24 62 74

Devon Ryan

unread,
Dec 10, 2019, 2:51:41 PM12/10/19
to Iddir Yasmine, deep...@googlegroups.com
Hi Yasmine,

bamCoverage has a few options for normalization, of which usually RPKM
or RPGC (aka 1x normalization) are used. Of those, I prefer RPGC,
since it normalizes the sample to an average 1x coverage. You can also
have it produce an unnormalized track, which is handy in some cases.

Regarding computeMatrix, you're correct that it assumes that your
bigWig files were created in a way to make them comparable in some
way, since it performs no normalization. This is useful, since you
might combine RNA-seq, ChIP-seq and methylation data in the same
matrix and all of those will necessarily need to be on very different
scales. The only sort of normalization that computeMatrix can do is if
you select `scale-regions`, which causes it to normalize regions to
the same length. In short, computeMatrix iterates over the regions in
your BED/GTF file(s) and, for each extracts the signal from your
bigWig files and averages. If requested (e.g., with `scale-regions`),
it will scale the length of these regions so that they're all the same
length (e.g., by interpolating) and then save the result in a big
compressed text file with per-bin values.

plotHeatmap mostly just plots the result. It can perform some
clustering and such as well.

Devon
--
Devon Ryan, Ph.D.
Email: dpr...@dpryan.com
Data Manager/Bioinformatician
Max Planck Institute of Immunobiology and Epigenetics
Stübeweg 51
79108 Freiburg
Germany
> --
> You received this message because you are subscribed to the Google Groups "deepTools" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to deeptools+...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/deeptools/2b036507b094499f94a2d443884092cf%40curie.fr.
Reply all
Reply to author
Forward
0 new messages