New to Bioinformatics

[Aholton]

Hello, I am new to Bioinformatics and it's software. I was wondering if I could ask a sample question and see what discussions came up so I could learn a bit more about it.

Let's say I have 2 RNA sequences from humans in FASTQ format and wish to do a comparison of the two through gene expression to figure out what the differences are between the two sequences.

I have looked at programs like the Tuxedo Suite, Galaxy, and BLAST. I would like to use our own 12 core processor to do the calculations, some website interface could be used as well.

Thank you for all your help!

PS: Is there anyway to get as specific as what membrane proteins are expressed in one but not the other?

[Krish_143]

first you have to find which membrane proteins i.e which is mem prot or not and specific to type .. later you have to find gene expression which one was expressed or not. then by combing these two you can get result whether type of membrane proteins expressed are not.. it is one way i know.

[Krish_143]

you have to change subject ( what you want ) and post again then you can get more replies.. i guess.

[Aholton]

Thanks for the fast response, think I'll edit the title and try again. Mainly just wanted to know which program would be able to compare the two sequences and tell me what was different about them.

[Krish_143]

do you have Fastq format (raw reads) or fasta format(sequence).. ?

[Aholton]

They are in Fastq format, I ran them through Clean_reads to cut out some of the lesser quality parts. First time doing all this so hopefully I'm on the right track.

Currently trying to get the Tuxedo Suite (Tophat, bowtie, etc.) on our server and see if I can't get some results from there.

Kinda dove head first into the project. hahaha

[Krish_143]

ok thats fine.. your correct.. after that finally for expression you have to use cummeRbund, EdgeR or any other..

for finding whether what type of membrane protein. you can paste fasta sequence http://www.imtech.res.in/raghava/pslpred/

mix this two results.

[Aholton]

Ok so if I was to write a guide for this, what is the Tophat/Bowtie part actually doing to the sequence? Is it lining them up to where they are easier to read?

[Krish_143]

simply Tophat is a tool used for mapping the short reads against the genome or transcriptiome. bowtie and bowtie2 are internally used in tophat for old and new versions respectively.

Yeah it was easy you can search in google scholar for articles you can understand after reading.
how to use why to use specifically..

[Aholton]

Thanks! I tried to Google it to understand more, but I was quickly lost in all the jargon that was being used.

[billstevens]

Hey, I know what its like.

I did what you are doing a few months ago. Let me try and outline it for you:

So Bowtie will take your reads and map them to your genome. Eukaryotic proteins use splicing, so you need a program that will figure out where your junctions, etc. are. That's where Tophat comes in. It takes all the reads that didn't map contiguously to the genome and figures out where it was spliced. At the end of this, you'll have a file called acceptedhits.bam. You are going to want to run Samtools on this, "samtools flagstat" which will give you your stats. It will tell you how many reads you have, and how many reads mapped. Obviously, the higher the number, the better.

You are then going to run Cufflinks on your hits.

Haha, on second thought, you can get almost all of this by reading the nature protocol paper by trapnell et al.

http://www.nature.com/nprot/journal/v7/n3/full/nprot.2012.016.html

Enjoy

[Aholton]

Thank you for the quick explanation, I'll be scouring that paper during lunch today. Hopefully it can put some things into perspective for me.