[BioC] Question about Diffbind input control

[Rory Stark]

Actually DiffBind passes only positive integers to DESeq, and the
normalization step is done by DESeq itself. If you also use edgeR, edgeR
does a separate normalization, with each package using the values they
normalized directly. You can see the (non-integer) normalized values after
an analysis us setting bCounts=TRUE in a call to dba.report.

-Rory

On 11/09/2013 17:52, "Anna Poetsch" <Anna.Poetsch at cancer.org.uk> wrote:

>Are the count values then also rounded to integers again?
>________________________________________
>From: Rory Stark [Rory.Stark at cruk.cam.ac.uk]
>Sent: Wednesday, September 11, 2013 5:50 PM
>To: Anna Poetsch
>Cc: bioconductor at r-project.org
>Subject: Re: Question about Diffbind input control
>
>Hi Anna-
>
>Before running DESeq, DiffBind sets all counts to a minimum value of 1.
>Not sure if that is in the documentation anywhere!
>
>Cheers-
>Rory
>
>On 11/09/2013 17:47, "Anna Poetsch" <Anna.Poetsch at cancer.org.uk> wrote:
>
>>Dear Rory,
>>
>>I attended the class about ChIP-Seq data analysis at EBI recently, where
>>you introduced us to the Diffbind package. There is still something that
>>I do not quite understand. It is about the correction for Input reads:
>>How is the correction for input done? Are the input counts just
>>subtracted after normalisation? If so, how does the package deal with
>>resulting negative counts, which would not be accepted by DESeq? Since
>>there is a normalisation step in between, the counts would not be
>>integers anymore, but DESeq needs integers. How is this solved?
>>I really tried to solve these questions looking at all the documentation
>>I could find and tried but failed to understand the source code. It would
>>be really nice, if you could help me with this.
>>
>>Kind regards,
>>Anna Poetsch
>
>
>NOTICE AND DISCLAIMER
>This e-mail (including any attachments) is intended for the above-named
>person(s). If you are not the intended recipient, notify the sender
>immediately, delete this email from your system and do not disclose or
>use for any purpose.
>
>We may monitor all incoming and outgoing emails in line with current
>legislation. We have taken steps to ensure that this email and
>attachments are free from any virus, but it remains your responsibility
>to ensure that viruses do not adversely affect you.
>Cancer Research UK
>Registered charity in England and Wales (1089464), Scotland (SC041666)
>and the Isle of Man (1103)
>A company limited by guarantee.  Registered company in England and Wales
>(4325234) and the Isle of Man (5713F).
>Registered Office Address: Angel Building, 407 St John Street, London
>EC1V 4AD.