Why is it necessary to visualize the RNA-seq data in a genome browser?

pinki999 · 04-06-2011, 01:04 AM

Hi,

Its a silly question but Why is it necessary to visualize the RNA-seq data in a genome browser?

schmima · 04-06-2011, 02:42 AM

hm - I don't think that there's a rule forcing you to do so

but it can definitely help you to get an idea of the data and to quickly check certain things.

Let's say you would like to know if the coverage over the transcripts is more or less uniform or not - having a look at it with genome browser will take you five minutes to get an idea if it may be uniform or absolutely not. Writing a script will take you definitely longer...

by the way - most of the people don't like to interpret thousands of numbers - pictures are generally preferred. So if you find a trend in your analysis, most readers will be happy if you can illustrate it with an example...

maubp · 04-06-2011, 08:13 AM

It is also useful to check gene models by eye - they should match up with the RNA-seq coverage information (e.g. intron/exon boundaries, or paired reads should all be within one transcript).

Azazel · 04-06-2011, 11:51 PM

I wouldn't say it's necessary. Many researchers like to take an actual look at the data to get ideas, as above posters wrote already; I admit, I also do that.

But this can be misleading and you can jump to wrong conclusions, and then you try to prove "what you see" using rigorous analysis, while actually it's not there.

There are many subtle issues with visualization of RNA-Seq/HT transcriptome profiling (pixel width, thresholds when tags appear/dissapear, coloring of tags which are smaller than one pixel, multiple tags within the same pixel while being zoomed out, etc, etc) which can give you wrong ideas when looking at the visualization.

vd4mindia · 08-05-2013, 09:13 AM

HI,

Am having a query for uploading the custom tracks. I need to know that if I want to visualize the RNA-Seq data then I have to upload the bigwig file for the same right? So lets say I have 1 patient with tumor and the conditions are having center and periphery of tumor. So for that patient I have RNA -Seq data of 4 samples with a center of tumor, periphery of tumor , recurrent center of tumor and recurrent periphery. So I want to view the data in Genome Browser. in order to do that I should convert all the BAM files of all the four and create a bigwig files for each and then add them using add custom tracks in the browser right?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
How to visualize RPKM in a browser?	dmotti	Bioinformatics	4	08-28-2015 11:07 AM
Identify and visualize differentially expressed genes from RNA-Seq data?	mediator	Bioinformatics	7	05-09-2012 01:00 AM
Visualize pair-end RNA-Seq in UCSC browser	oscarluoinau	Bioinformatics	0	11-21-2011 02:24 AM
RNA-Seq: ENCODE whole-genome data in the UCSC genome browser (2011 update).	Newsbot!	Literature Watch	1	11-24-2010 02:08 PM
ChIP-Seq: Savant: Genome Browser for High Throughput Sequencing Data.	Newsbot!	Literature Watch	0	06-22-2010 03:00 AM

04-06-2011, 01:04 AM	#1
pinki999 Member Location: Germany Join Date: Oct 2010 Posts: 37	Why is it necessary to visualize the RNA-seq data in a genome browser? Hi, Its a silly question but Why is it necessary to visualize the RNA-seq data in a genome browser?

08-05-2013, 09:13 AM	#5
vd4mindia Member Location: Milan Join Date: May 2013 Posts: 40	Viewing the custom tracks in UCSC genome browser HI, Am having a query for uploading the custom tracks. I need to know that if I want to visualize the RNA-Seq data then I have to upload the bigwig file for the same right? So lets say I have 1 patient with tumor and the conditions are having center and periphery of tumor. So for that patient I have RNA -Seq data of 4 samples with a center of tumor, periphery of tumor , recurrent center of tumor and recurrent periphery. So I want to view the data in Genome Browser. in order to do that I should convert all the BAM files of all the four and create a bigwig files for each and then add them using add custom tracks in the browser right?

04-06-2011, 02:42 AM	#2
schmima Member Location: Zürich Join Date: Apr 2010 Posts: 56	hm - I don't think that there's a rule forcing you to do so but it can definitely help you to get an idea of the data and to quickly check certain things. Let's say you would like to know if the coverage over the transcripts is more or less uniform or not - having a look at it with genome browser will take you five minutes to get an idea if it may be uniform or absolutely not. Writing a script will take you definitely longer... by the way - most of the people don't like to interpret thousands of numbers - pictures are generally preferred. So if you find a trend in your analysis, most readers will be happy if you can illustrate it with an example...

04-06-2011, 08:13 AM	#3
maubp Peter (Biopython etc) Location: Dundee, Scotland, UK Join Date: Jul 2009 Posts: 1,543	It is also useful to check gene models by eye - they should match up with the RNA-seq coverage information (e.g. intron/exon boundaries, or paired reads should all be within one transcript).

04-06-2011, 11:51 PM	#4
Azazel Member Location: Japan Join Date: Oct 2010 Posts: 52	I wouldn't say it's necessary. Many researchers like to take an actual look at the data to get ideas, as above posters wrote already; I admit, I also do that. But this can be misleading and you can jump to wrong conclusions, and then you try to prove "what you see" using rigorous analysis, while actually it's not there. There are many subtle issues with visualization of RNA-Seq/HT transcriptome profiling (pixel width, thresholds when tags appear/dissapear, coloring of tags which are smaller than one pixel, multiple tags within the same pixel while being zoomed out, etc, etc) which can give you wrong ideas when looking at the visualization.