Hello
I recently analysed stranded illumina trueseq paired-end RNA-seq data. My data is 1+-,1-+,2++,2--, meaning the second_in_pair should correspond to the original strand of the fragment.
What I observe there is that the majority (~90%) of fragments seem to origin from the expected strand (the one with the annotated gene. But a minority (~10%) are assigned to the opposite strand. I am very positive that this is not true anitsense transcription since the coverage follows the same coverage pattern which nicely fits the exon-intron annotation.
Therefore my questions:
+ have some of you experienced similar issues?
+ what is the expected strand information specificity of a stranded illumina trueseq paired-end RNA-seq?
+ does someone know how this antisense shadow are introduced during library perp?
+ is someone aware of resources where such issues are discussed?
any help/suggestions/comments are highly welcomed.
thanks in advance
I recently analysed stranded illumina trueseq paired-end RNA-seq data. My data is 1+-,1-+,2++,2--, meaning the second_in_pair should correspond to the original strand of the fragment.
What I observe there is that the majority (~90%) of fragments seem to origin from the expected strand (the one with the annotated gene. But a minority (~10%) are assigned to the opposite strand. I am very positive that this is not true anitsense transcription since the coverage follows the same coverage pattern which nicely fits the exon-intron annotation.
Therefore my questions:
+ have some of you experienced similar issues?
+ what is the expected strand information specificity of a stranded illumina trueseq paired-end RNA-seq?
+ does someone know how this antisense shadow are introduced during library perp?
+ is someone aware of resources where such issues are discussed?
any help/suggestions/comments are highly welcomed.
thanks in advance