[mfahim]

smallRNA Analysis Pipeline

Hi Simon.
introducing RNAseq quantitation pipeline to seqmonk is great. Are you planning to add a pipeline for smallRNA soon?. If not how do I define probe for small RNAs.

Thanks
Fahim

[simonandrews]

The RNA-Seq quantitation has been in for a while, but we now make it easy to link out to DESeq and EdgeR without having to leave the program.

For smallRNAs you could still use the RNA-Seq pipeline but just change the feature type you want to quantitate to the appropriate small RNA category (miRNA for example). Given that small RNAs aren't generally spliced though you could just use a normal read count quantitation too.

There's also the small RNA QC plot which is a really useful way to start looking at your small RNA data.

[mfahim]

Thanks Simon---However

Thanks for the tip off.. Howerver, I am a bit worried about the cummulative distribution graph after percentile re-quantitation and match distribution quantitation. here is the snapshot.

[simonandrews]

Thanks for the tip off.. Howerver, I am a bit worried about the cummulative distribution graph after percentile re-quantitation and match distribution quantitation. here is the snapshot.

I can't see anything attached...

[mfahim]

Oh. previous snapshot was too big perhaps..here is another one. Cheers

[simonandrews]

Is this RNA-Seq or small RNA? What is the quantitation? Did you do the corresponding QC plot for the data type you're using - if so, what did that look like? You said you did two levels of additional normalisation, were they obviously needed? Did the distribution change much after applying them?

You seem to have a very large proportion of unmeasured probes which seems suspicious. Does the raw data look the way you'd expect?

[mfahim]

1. is this RNA-Seq or small RNA?
Ans.Yes, this is smallRNA seq.

2. What is the quantitation?
Ans. As you mentioned previously, I applied the normal RNAseq Quantitation pipeline.

3. Did you do the corresponding QC plot for the data type you're using - if Ans. so, what did that look like?

4. You said you did two levels of additional normalisation, were they obviously needed? Did the distribution change much after applying them?

Ans. Yes, as per your youtube instruction I went to check the plot and after application of the two normalization levels. This is how it appears (the shared picture), improved as compared to the previous plot but obviously still bad.

5. You seem to have a very large proportion of unmeasured probes which seems suspicious. Does the raw data look the way you'd expect?
The QC plot was having some error, I have sent the report.
I have no idea what is happening here.

[mfahim]

Suggestion for next version of SeqMonk

Hi Simon,
Like many others on this forum, I am a big fan of SeqMonk.
Here is a suggestion for next version (or may be I dont understand SeqMonk).
Yes, we can make customized Annotation Tracks BUT what if I have a specific list of genes involved in a specific process then how do I make annotation track.

I guess there is room for improvement here, an option may be included where I can upload a list of genes with a variety of fuctions (names or TAIRids) and then make a customized Annotation Track. If this is added to the SeqMonk next version, that will greatly help my analysis.

If there is already one then I am sorry for oversight.

I am working with flowering time in Arabidopsis and there are several pathways involved such as Vernalization pathway, Photoperiod Pathway, Ambient Temperature Signalling Pathway, Autonomous Pathway and some Biochemical Pathways, furthermore new genes are discovered on daily basis that add to the flowering time and the TAIR database is not regularly updated.

Your kind insight will be highly appreciated.

Best Regards
Fahim

[simonandrews]

Hi Simon,
Like many others on this forum, I am a big fan of SeqMonk.
Here is a suggestion for next version (or may be I dont understand SeqMonk).
Yes, we can make customized Annotation Tracks BUT what if I have a specific list of genes involved in a specific process then how do I make annotation track.

You can already make a track out of a subset of an existing track. In your case if you have a list of gene names then you could do:

Edit > Find named features

Paste your list of names into the box and select 'gene' as the features to search. It will then match your gene names against the names of the gene features and give you a list of hits. Once you have this you can press "Save All as Annotation Track" to turn the list of hits into a new track which you can then use for your downstream analysis.

Does that do what you want?

[mfahim]

Bingo!! Thank you Simon

for the Genelist..
That is exactly what I was looking for. Perfect.

More questions next week...

Cheers
Fahim

[mathew]

Question about Seqmonk annotations

I am using Seqmonk 30. I have ran Chip analysis and has annotations like CGI, TSS, mRNA, Promoter etec. I want to export coordinates with all the annotations in one file Is it possible how to get around.

Thanks

[simonandrews]

I am using Seqmonk 30. I have ran Chip analysis and has annotations like CGI, TSS, mRNA, Promoter etec. I want to export coordinates with all the annotations in one file Is it possible how to get around.

If you want the raw annotations which seqmonk loads then you can just go to the Genomes folder where you'll find a set of .dat files which are EMBL format sequence header files which contain all of the positions of the annotations you see when you load a core genome.

For individual tracks you could make probes over the features using the feature probe generator and then create an annotated probe report to get a simpler tab-delimited output with coordinates in. The only thing you wouldn't get by doing it this way would be the positions of any sub-features (ie exons within transcripts).

[mathew]

.dat file will give the genome coordinates of that particular gene. If I create a report that will be for one annotation i.e either TSS 2000(say) bp up down / mRNA/ CGI / intergenic and so on. What I want to do is to call peaks and annotate each peak with any region whether it is in enhancer, promoter, CGI, mRNA assigning all criteria in one output file.
So I am looking for a file outlined below-

chr1 100 1000 geneXX cgi 2000bp enhancer Tss 1000 bp RPKM 1 RPKM2 Diff

[simonandrews]

We've just released seqmonk v0.31.0 onto the project web site. We've added some new options for analysing RNA-Seq and BS-Seq data and have tidied up the user interface for cases where you have long names for your data stores or probe lists.

http://www.bioinformatics.babraham.a...jects/seqmonk/

[simonandrews]

SeqMonk v0.32.1 has just been released. This fixes a bug which meant that on some systems the dialog which showed the progress of R scripts could hang. We've also taken the opportunity to improve the reporting of R crashes so that the full R log is now also sent alongside the crash report, which should help to sort out any remaining R issues.

We also improved the line graph, adding the ability to click on a line to see which probe it came from, and double click on it to go to that probe in the chromosome view.