An anxious Msc student doing a splice-junction tools research project

[kanewong]

Hi all,

Nice to meet you all. I was so glad when I first knew of this amazing forum.

I am a msc student from HK. I started my research project related to splice-junction tools a month ago, which is absolutely beyond my ability cause I am not familiar with Linux at all... ( From secondary school to undergraduate I was studying biochem)

I will need to hand in my paper at mid of May........, hopefully can learn important skills and knowledge related to splice-junction tools from friends in this forum. at an impossible rate

Is there any friends working on splice-junction tools here? It will be my great pleasure to be your friend. Most of my real life friends are having biochem background, so I am lonely in doing bioinformatics work. This forum will likely be my most important sorce of help in my research project.

[rjohnp]

Hi,

I know that feeling, I've gone straight from being a cellular biologist to doing my PhD in pure NGS data analysis, minimal Linux or coding experience prior.

It would help to know what data you are analysing splicing on? E.g. do you have RNA-Seq data, or just looking at reference genomes? What are you actually trying to learn from your data? The tools will be different depending on this.

[kanewong]

Hi,

I know that feeling, I've gone straight from being a cellular biologist to doing my PhD in pure NGS data analysis, minimal Linux or coding experience prior.

It would help to know what data you are analysing splicing on? E.g. do you have RNA-Seq data, or just looking at reference genomes? What are you actually trying to learn from your data? The tools will be different depending on this.

Hi rjohnp, yap so many pain entering a dry lab after working so many years in a wet lab

I am using a set of Arabidopsis Thaliana RNA-seq data from here

http://www.ncbi.nlm.nih.gov/sra/SRX103665

In short, my research requires me to run the same set of data using 3 different splice-junction detection tools, and compare the results produced by these 3 tools. I will use a tool to trim the data into different read length and see how these 3 tools respond in a different way(if any) to different read length. Sounds simple, but several problems are nightmares for me:

1. I still don't know how to use a tool to compare the results(splice-junction) of different tools

2. I haven't tried using a tool to trim a RNA-seq file before

[shi]

Hi,

There are around 4 or 5 popular junction detectors in this field. We recently compared few of them, which may help you to start your project. Here is the link to our evaluation results:

http://nar.oxfordjournals.org/conten...kt214.abstract

I do not quite understand why you want to try to trim reads for the junction detection. Junction detectors work better with longer reads because longer reads are more likely to span exons. Given that the popular sequencers (eg Illumina HiSeq) now generate 100bp or longer reads, you should not really look at reads much shorter than that.

Hope this helps.

Cheers
Wei

[kanewong]

Hi,



I do not quite understand why you want to try to trim reads for the junction detection.

Hi, actually I feel strange too...My supervisor gave me a project and its aim is to compare the performance of splice-junctino detection tools. I am too new to this field, and finally I could just think of this idea.

I think probably my major focus now is to determine which tool can tolerate a shorter read length of input data.