Comparing Multiple Species RNA seq data

Question

[Lmira zand]

I want to compare Multiple vertebrate species RNA seq data(more than 11 species) in the same organ to realize the transcriptome evolution between them in the target Organ, I have the alignment files of these species, I must point out that there are different replicates per species(e.g. 4 rep for Bos, 6rep for Mus, 1 rep for Macaca) I'm completely new to this field and I need bits of advice to make a most accurate pipeline for my analysis. Any advice is valuable to me. Thanks a lot in advance.

[GenoMax]

I want to compare Multiple vertebrate species RNA seq data(more than 11 species) in the same organ to realize the transcriptome evolution between them in the target Organ

Has this been done before? I am not sure how one would conclude something about evolution by looking at expression levels.

[Lmira zand]

It has been done in maximum 6 species and other organs than my target organ, it is sufficient to get genes which differentially expressed between them to get the change of expression levels between these species in this organ.

[GenoMax]

If there is a publication can you link it here?

Please use ADD COMMENT/ADD REPLY when responding to existing posts to keep threads logically organized. SUBMIT ANSWER should be used for new answers to original question.

[Lmira zand]

https://onlinelibrary.wiley.com/doi/full/10.1111/j.1365-313X.2012.05005.x

[GenoMax]

If I am reading the publication right then the evolution part was analyzed independent of the expression (finding orthologous groups). The groups were then used for co-expression analysis.

[Lmira zand]

I am confused to separate these species to groups or not because I don't know how to select these species in groups; according to the number of replicates or types, and I need an exact pipeline for analyzing.

[GenoMax]

You could try what the authors of the paper above did:

To evaluate conservation of expression across species, we first defined orthologous and/or close paralogous groups using predicted protein‐coding genes and the OrthoMCL algorithm

[Lmira zand]

Thank you so much I will try your suggestion and start with identifying paralogous and orthologous genes.

Answer 1

[johnsonnathant]

Over the year, there have been several papers asking a similar question, but the definite first step in the analysis is determining what is the 'same gene' usually done via orthologues or paralogues.

Here are some examples on cross species/genus/organism comparisons:

https://www.pnas.org/content/111/48/17224

https://www.nature.com/articles/sdata2017185

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4654840/

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3091293/

[Lmira zand]

Thank you, Johnson, I will check these papers. You are right I must get those genes first.