Seeing 3nt periodicity

[gagliajm]

Hi folks,

Wow this site is exactly what I need. Thank you very much for making this forum available. I am a newbie to the ribosome profiling world. I am a part time grad student and I have recently sequenced my first ribosome profile for a viral infected human cell. Now I am trying my best not to have to call in a card carrying bioinformaticist to fully analyze my data but rather do it myself. And I may fail miserably.

I have use CLCbio and Galaxy tools to some degree. My first goal is to take my data, map it and see if I can detect the "canonical" 3 nucleotide periodicity that many groups can see from their profile.

Any tips or warnings?


My sincere apologies for questions that are dumb or too general. I am an expert in the wet lab but not so much when it comes to bioinformatics. I am trying out RiboGalaxy for the first time today so I will see how that goes.

Thanks

Jason

[First citizen of GWIPS]

Hi Jason, welcome to the forum. Incidentally we just opened RiboGalaxy (http://ribogalaxy.ucc.ie) to the public. You have a chance to become the first user. You may or may not see periodicity depending on the protocol. I would say that a better first test would be to count read density in 5'leaders, CDS and 3' UTRs. For the control (mRNAseq) the density should be uniform, for riboseq you should expect CDS>5'leaders>3'UTRs

[gagliajm]

(09-Jan-2015 08:40 PM)First citizen of GWIPS Wrote:  Hi Jason, welcome to the forum. Incidentally we just opened RiboGalaxy (http://ribogalaxy.ucc.ie) to the public. You have a chance to become the first user. You may or may not see periodicity depending on the protocol. I would say that a better first test would be to count read density in 5'leaders, CDS and 3' UTRs. For the control (mRNAseq) the density should be uniform, for riboseq you should expect CDS>5'leaders>3'UTRs

Thank you for your response. Sorry for the delay...

Since I am new to all of this I took some time and read about 15 papers on ribosome profiling to get a sense of the analysis tools folks use to extract meaningful information from the data.

After my grad exam I will have more time to devote to the analysis of my data.

Thanks again

Jason

[yyu]

thanks for the answer, I have a following question -
Is there a script in Riboseq that can quantify the read in 5'UTR vs CDS vs 3'UTR? It's obvious in my data when I look at browser, but it's nice to have the number.

Thanks a lot!


Yingpu

[vimalkumarvelayudhan]

Hi Yingpu,
You can use ribocount (available under RiboSeq analysis -> Riboplot) to get
read counts for all transcripts in an alignment (sorted BAM). You can choose
which read counts to output:

1. The entire transcript
2. 5' region of the longest ORF
3. 3' region of the longest ORF

CDS counts are not available at the moment. You could probably work this out
from the outputs above.

Regards,
Vimal

[yyu]

Hey Vimal,

Thanks for your reply!
I am a little confused here. Can I use the BAM from transcriptome mapping for ribocount? I tried, full length seems ok but both 5' & 3' UTR is empty.
I trying another library in case it's library specific but I really doubt it.
Anyway, I'm still sharing my history with Ribogalaxy and feel free to check whenever you need to. Thanks a lot again for you help!


Yingpu