I have developed a RSEM analysis in order to get a differential expression matrix.
I have obtained a matrix of a list of genes with this kind of ID: c48895_g2.
I can't realised a enrichment of GO terms with these IDs. How can I convert these IDs to gene names?
If I understand correctly, you need to map RSEM internal IDs to something like HGNC gene symbols or transcript ids. The easiest way to do this is abandon the RSEM ids and use the ranges. You'll need to overlap the RSEM ranges with ranges of known genes. The GenomicFeatures and GenomicRanges libraries in R are great for annotating. There are plenty of tutorials on how to find overlapping ranges using the functions in these libraries.
Alternatively you can use PostGreSQL and an R tree index if you like databases, or maybe BEDtools.
Dear Pedro
How small the dataset may be, what you need is a simple ID mapper. You can do it using NCBI's Batch Entrez or Uniprot's ID mapper or HGNC, if it were human gene ids.
Best
Prash
What type of ID is "c48895_g2"? You can easily convert gene IDs using online tools DAVID (http://david.abcc.ncifcrf.gov) or Uniprot (http://www.uniprot.org/mapping/), or if the data set is particularly large you can download flatfiles from NCBI and run a quick little script to match them.
Yes Jean. I followed the Trinity/ RSEM/ EdgeR pipeline and have produced two list with genes and isoforms. These are the names, and I am not able to extract the sequence of thise genes and isoforms from the Trinity.fasta.
Is there such a way/ program to do this task?
Thank you!!
You will need Trinotate (http://trinotate.github.io/ ) to complete your pipeline. Among others, it involves Blasting all your contigs to find which genes they correspond to (or that are 'similar' to your assembled contigs).
http://trinotate.github.io/
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Bioinformatics
- Bioinformatic Tools