Request: Supporting BamFile and BamFileList as `bamFiles` arg

[Peter Hickey]

Would you consider supporting BamFile and BamFileList objects (from the Rsamtools package), in addition to character vectors, as the bamFiles argument to functions in this package? This actually almost works with the current code, so I don't think it will require many changes to be made.

A concrete example of where this is useful: I have a bunch of ATAC-seq BAM files where the (.bam, .bai) pair are of the form (file.bam, file.bai). Right now, I have to create a symbolic link from file.bai to file.bam.bai because this is the format ATACseqQC expects. With this suggested change, I could instead do bfl <- Rsamtools::BamFileList("file1.bam", "file2.bam"), which automatically picks up the file1.bai and file2.bai indices, and all the Rsamtools and GenomicAlignments functions used by ATACseqQC would immediately work with these BamFile and BamFileList objects.

[JIANHONG OU]

Thank you for trying ATACseqQC. Yes, you are correct. It should be fine with the bamFile object. But for BamFileList is still not supported yet. Currently, I am still working on how to decrease the memory cost of the program. After that, BamFileList will be supported. And you can always support index file as a parameter when you read in bam files. Please have a try. Jianhong

…

On Fri, Apr 6, 2018 at 3:56 PM, Peter Hickey ***@***.***> wrote: Would you consider supporting *BamFile* and *BamFileList* objects (from the *Rsamtools* package), in addition to *character* vectors, as the bamFiles argument to functions in this package? This actually almost works with the current code, so I don't think it will require many changes to be made. A concrete example of where this is useful: I have a bunch of ATAC-seq BAM files where the (.bam, .bai) pair are of the form (file.bam, file.bai). Right now, I have to create a symbolic link from file.bai to file.bam.bai because this is the format *ATACseqQC* expects. With this suggested change, I could instead do bfl <- Rsamtools::BamFileList("file1.bam", "file2.bam"), which automatically picks up the file1.bai and file2.bai indices, and all the *Rsamtools* and *GenomicAlignments* functions used by *ATACseqQC* would immediately work with these *BamFile* and *BamFileList* objects. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#1>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AFYSA9uBhsbpoZQrdwntWHQdRF2fzkqTks5tl8hvgaJpZM4TKnwj> .

-- Yours sincerely, Jianhong Ou

[Peter Hickey]

Thanks, Jianhong. How can I specify the index file to a function? For example, I can't see how to do this for the fragSizeDist() function.

For what it's worth, switching to BamFile and BamFileList may enable some additional control over memory usage via the yieldSize argument to the BamFile() and BamFileList() constructors.

[JIANHONG OU]

Thank you for your suggestion. fragSizeDist will accept index in next build. You can git clone it now. Yes yieldSize should help to reduce the memory cost. However, it is still hard for me to use this parameter for all functions such as to get the uniquely mapped reads.

…

On Fri, Apr 6, 2018 at 4:36 PM, Peter Hickey ***@***.***> wrote: Thanks, Jianhong. How can I specify the index file to a function? For example, I can't see how to do this for the fragSizeDist() function. For what it's worth, switching to *BamFile* and *BamFileList* may enable some additional control over memory usage via the yieldSize argument to the BamFile() and BamFileList() constructors. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#1 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AFYSA3Kjhk8kZjiBzhphkYo3Cd9bI37eks5tl9G0gaJpZM4TKnwj> .

-- Yours sincerely, Jianhong Ou