Tweet
Hi all,
I am currently working with my first small RNA dataset (generated on Illumina GAII). The preprocessed dataset contains about 16 million reads. I have figured out how to use the various tools in Galaxy to do preprocessing steps such as trimming, clipping, mapping, etc, but have a question about classifying small RNA populations. I'm comfortable with programming so suggestions for homegrown solutions are welcomed.
In C. elegans, there are a number of classes of small RNAs such as 21U, 22G, 26nt, etc. I would like to separate the reads of these classes into their own Fastq files, however I know that there will be degradation products (some 19nt and 20nt RNAs will be degradation products of of the 21U RNAs) that I would also like to classify into these classes. Is there an established method for doing so? I feel like pairwise alignments against all other reads in that category would be too computationally intensive, even if only performed on the first n nucleotides of each read. Is it? Anyone have any other suggestions?
Thanks!!
I am currently working with my first small RNA dataset (generated on Illumina GAII). The preprocessed dataset contains about 16 million reads. I have figured out how to use the various tools in Galaxy to do preprocessing steps such as trimming, clipping, mapping, etc, but have a question about classifying small RNA populations. I'm comfortable with programming so suggestions for homegrown solutions are welcomed.
In C. elegans, there are a number of classes of small RNAs such as 21U, 22G, 26nt, etc. I would like to separate the reads of these classes into their own Fastq files, however I know that there will be degradation products (some 19nt and 20nt RNAs will be degradation products of of the 21U RNAs) that I would also like to classify into these classes. Is there an established method for doing so? I feel like pairwise alignments against all other reads in that category would be too computationally intensive, even if only performed on the first n nucleotides of each read. Is it? Anyone have any other suggestions?
Thanks!!