Following Janet's example, I would also like to propose an upgrade to ScanBamParam:
It would be great if we could tell ScanBamPram that we want to load only the reads that passed the vendor's quality filter. In other words, the functionality I am suggesting is analogous to the filter in readAligned() from the ShortRead library. With the new release of Illumina sequencing reagents (version 3) you get 200 million reads per lane from the HiSeq 2000. In my view, with samples that big becoming popular, any investment in "read in" efficiency is a good investment. I would be happy to provide a sample BAM for those interested in addressing this suggestion. It is also my humble opinion that we should start considering parallelisation for reading in. I hope that I am not just wishing too much. Thank you, Ivan _______________________________________________ Bioc-sig-sequencing mailing list Bioc-sig-sequencing@r-project.org https://stat.ethz.ch/mailman/listinfo/bioc-sig-sequencing