Use an annotated transcript in STAR but no .gtf file available

Question

[luzglongoria]

Hi there,

I'm going to start with my RNA-seq analyses using STAR for mapping. I have no reference genome perse , what I have is an annotated transcript from a previous study. I made this annotated transcript by using Trinity (de novo assembling) and I wonder if it is possible to run STAR with this Trinity file as a reference genome.

This Trinity file is annotated but I don't have a .GFT file to include in the Genome Indexing Parameters for STAR. I know this option is only used for –runMode genomeGenerate but it's going to be the first time for me and I'm not sure how to do it.

I just run my read in a normal way with this Trinity file or I should do some thing before?

Any help is more than welcome.

Best,

Luz

Answer 1

[Dave Carlson]

STAR is a spice-aware aligner that is designed for mapping to a genome assembly, not a transcriptome.

If all you have is the transcriptome assembly, I would recommend using a different tool, depending on what your downstream goals are. For example, you could use RSEM for mapping + quantification. Or Salmon or Kallisto for quantification.

Trinity comes with several scripts that make this process easier. See the following:

https://github.com/trinityrnaseq/trinityrnaseq/wiki/Trinity-Transcript-Quantification

[luzglongoria]

Thank you so much for your answer.

My goal if to filter my dataset since I’m working with parasites infecting another organism (mix reads). I would like to use this transcriptome assembly in order to filter and then (and after other filtering process) quantify with Trinity tools , as you suggested.

Maybe I could use other tools as Hisat2?

[Dave Carlson]

I believe Hisat2 is also designed for mapping RNA-Seq reads to a genome instead of a transcriptome. However, if you use RSEM, you can map your reads to your transcriptome assembly using bowtie2.

[luzglongoria]

Great ! Thank you so much.