[BioC] CRISPRseek: findgRNAs bug?

[Zhu, Lihua (Julie)]

Shamit,

FYI, if you use offTargetAnalysis ( chromToSearch = "") instead of findgRNAs
function, you will have all the gRNAs annotated with restriction enzyme
recognition patterns.

I did not get to see you in the Bioc2014 conference in Boston. If you are
interested in using it for off target search and cleavage prediction, here
is the slide and hands-on exercises at the Bioconductor course website.
There are workshop materials for other packages that might be of your
interest.

http://master.bioconductor.org/help/course-materials/2014/BioC2014/

Best regards,

Julie


On 7/11/14 3:56 PM, "Lihua Julie Zhu" <julie.zhu at umassmed.edu> wrote:

> Shamit,
> 
> You are very welcome! Thanks for the feedback!
> 
> BTW, when you have additional question, please cc bioconductor at r-project.org
> so others also benefit.
> 
> Best regards,
> 
> Julie
> 
> 
> On 7/11/14 10:32 AM, "Shamit Soneji" <shamit.soneji at med.lu.se> wrote:
> 
>> Ok great, thanks very much, and thanks for seeing to it so quickly.
>> 
>> Best,
>> 
>> Shamit
>> 
>> On Fri, 2014-07-11 at 10:29 -0400, Umassmed wrote:
>>> I have committed the changes. It should appear soon in the bioc site. If you
>>> are in a rush, you can obtain it from
>>> http://master.bioconductor.org/developers/how-to/source-control/
>>> 
>>> FYI, I am attending a conference till Tuesday so will not be as responsive
>>> as
>>> usual.
>>> 
>>> Best regards,
>>> 
>>> Julie
>>> 
>>> Sent from my iPhone
>>> 
>>>> On Jul 11, 2014, at 3:42 AM, Shamit Soneji <shamit.soneji at med.lu.se> wrote:
>>>> 
>>>> Thanks Julie,
>>>> 
>>>> When will the package be updated to the BioC site? If possible a copy of
>>>> the amended function would be good.
>>>> 
>>>> Thanks,
>>>> 
>>>> Shamit
>>>> 
>>>>> On Thu, 2014-07-10 at 21:18 +0000, Zhu, Lihua (Julie) wrote:
>>>>> Shamit,
>>>>> 
>>>>> The bug has been fixed and you could also directly input a DNAStringSet
>>>>> object as inputFilePath.
>>>>> 
>>>>> Please let me know if you encounter any problem.
>>>>> 
>>>>> Many thanks for catching and reporting this bug!
>>>>> 
>>>>> Best regards,
>>>>> 
>>>>> Julie
>>>>> 
>>>>> 
>>>>>> On 7/10/14 10:56 AM, "Shamit Soneji" <shamit.soneji at med.lu.se> wrote:
>>>>>> 
>>>>>> Hi Julie,
>>>>>> 
>>>>>> My name is Shamit, I'm a bioinformatican at Lund University Sweden who
>>>>>> has been asked by a colleague to find CRISPR sites for promoters in the
>>>>>> human genome. I was looking at your package, specifically the findgRNAs
>>>>>> function.
>>>>>> 
>>>>>> I modified it slightly so that it would take a preloaded DNAStringSet
>>>>>> rather than having to load it from a file, but I've noticed the number
>>>>>> of found gRNAs is always from the last sequence in the input fasta.
>>>>>> 
>>>>>> In short the function you have is:
>>>>>> 
>>>>>> findgRNAs <- function(inputFilePath, .......){
>>>>>> 
>>>>>> subjects <- readDNAStringSet(inputFilePath, format
>>>>>> 
>>>>>> 
>>>>>> for(i in 1:length(subjects){
>>>>>> #does stuff here
>>>>>> }
>>>>>> 
>>>>>> 
>>>>>> all.gRNAs # the returned object
>>>>>> 
>>>>>> }
>>>>>> 
>>>>>> 
>>>>>> Maybe I'm not understanding how this works, but "all.gRNAs" is not
>>>>>> initialised outside the scope of the for-loop, and it is not appended to
>>>>>> in the loop so it never "grows".
>>>>>> 
>>>>>> If I run the function with the first 3 sequences, ie:
>>>>>> 
>>>>>> crtest <- findgRNAs(promoters[1:3],.....)
>>>>>> 
>>>>>> length(crtest)
>>>>>>> 60
>>>>>> 
>>>>>> 
>>>>>> Now...
>>>>>> 
>>>>>> if I run each one individually:
>>>>>> 
>>>>>> crtest <- findgRNAs(promoters[1],.....)
>>>>>> 
>>>>>> length(crtest)
>>>>>>> 61
>>>>>> 
>>>>>> crtest <- findgRNAs(promoters[2],.....)
>>>>>> 
>>>>>> length(crtest)
>>>>>>> 56
>>>>>> 
>>>>>> crtest <- findgRNAs(promoters[3],.....)
>>>>>> 
>>>>>> length(crtest)
>>>>>>> 60 #which is the same as the first run trying 3 sequences at once.
>>>>>> 
>>>>>> 
>>>>>> Am I right in saying that I should get 61+56+60=177 gRNAs when I run the
>>>>>> function for promoters[1:3]? I just find the fact all.gRNAs doesn't grow
>>>>>> in the loop.
>>>>>> 
>>>>>> If you could please tell me whether this is a bug, or whether I'm being
>>>>>> stupid I would appreciate it!
>>>>>> 
>>>>>> Best,
>>>>>> 
>>>>>> Shamit
>>>> 
>> 
>