background for RNAseq
Does anyone think the peaks in non-coding region could be some bias or random events, rather than some potential ncRNA? BTW, the organism is prokaryotic. Thanks!
-
Last edited by nivea; 12-14-2011, 03:11 PM. -
Google "pervasive transcription."
I think there is far more "noise" in transcription than people give credit too. Of course there is far from any consensus on the matter.
RNA Polymerase II can have a lot of false starts, producing transcripts that may or may not be functional:
-
Thanks chadn737. But my genome is a bacterial genome... So I'm wondering except for the situation of ncRNA, can these peaks found in the non-transcribed regions are generated by some sequencing bias or just random events?Comment
-
It is extremely unlikely that the sequencer randomly comes up with a sequence that actually maps to your genome unless this sequence was actually in the sample. So, this is definitely not a technical proble -- the non-coding sequences were in your sample, and so they were also transcribed by your cells. (Alternatively, it could be genomic DNA that sneaked into your RNA sample.)
Why do you think that it makes a difference that your sample is prokaryotic?Comment
-
Thanks Simon. But based on my understanding, the "pervasive transcription" in the mammalian genomes include the transcribed regions outside coding gene, anti-sense RNAs and other non-coding RNAs. In other words, some intergenic regions just purely transcribed without turning into any functional product (non-coding RNAs, antisense RNAs). In the case of prokaryotes, I don't know for sure...Originally posted by Simon Anders View Post
Like you said, the small peaks in the intergenic regions could be genomic DNA. Or it could also be novel ncRNAs with low expression level. If it's the previous situation, I can consider it as a noise in my background, and include it into my negative set for training; but if it's the latter case, I should definitely exclude the peak region to train my negative set.
BTW, when I mapped the reads to the reference genome, I only kept the uniquely mapped reads with tolerance of 3 mismatches. The read length is 100nt. Therefore, I don't think the small peaks are mapping artifacts...Comment
-
Would gDNA contamination actually make a peak? I wouldn't think so...Comment
-
Exactly. So it is actual RNA. Not that surprising, I think. After all, nobody says it is functional.Originally posted by biznatch View PostComment
-
Thanks chadn737, I think the paper you forwarded are very helpfulOriginally posted by chadn737 View PostGoogle "pervasive transcription."
I think there is far more "noise" in transcription than people give credit too. Of course there is far from any consensus on the matter.
RNA Polymerase II can have a lot of false starts, producing transcripts that may or may not be functional:
http://www.nature.com/nsmb/journal/v...b0207-103.html
Comment
-
Researchers using sequencing and genomics tools often have to make trade-offs. They can choose between speed or scale, short reads or long-range information, or targeted panels or a view of the whole transcriptome. New technologies that have been released this year are built to address those tough choices.
We asked six companies the same four questions to learn about their latest products. The new technologies bring a lot to the table, including rethinking sequencing...
Comment