I have RNA seq data which I have mapped using tophat.
but the alignmnet of these sequences is very poor . How many aligned reads should be there for selection of RNA sequneces.can it be more than 1 million or its better to take more than 2 million aligned reads?
I think it would be helpful if you gave us a bit more information. Which samples are these (cells/tissue?), which organism (human/plant/mouse?)? How were they prepared (RNA isolation etc)? Which RNA-Seq machine was used? Is this mRNA, miRNA? What kind of reads (single-end/paired-end) and how large library size do you have? Have you run FastQC before alignment?
I agree with Sindre, more information would be very helpful.
However, one quick way to determine if your soft ware settings are correct is to check what proportion of your raw reads align to the reference genome. In my experience this should be above 80%. Do you have a figure for this?
Rule of a thumb: 100M reads per treatment group for vertebrates to have a good resolution of counts on genes and see splicing variants.
Still - with single millions you may see "something" but loose many genes in the black hole of false negatives.
You may switch from tophat to star or subjunc for the sake of speed (order of magnitude better).
Hi Ahsan
depending on the organisms with which you are working you might need several millions of reads to perform a proper analysis. For bacteria it is usual to have about 5 millions whereas for eukaryotes the minimum is normally considered to be around 30 millions. If you have doubts about the results of tophat you could try different tools.
We have developed an online tool called AIR (https://transcriptomics.sequentiabiotech.com/) which let you perform a full RNA-seq differential expression analysis with just few clicks and with no previous bioinformatics knowledge required. The first 6 samples are free, if you want you could give it a try.
Kind regards
Riccado
- Badges
- Science topic
- Similar topics
- Biological Science
- Molecular Biology
- Biomolecules
- Nucleic Acids
- RNA