how many index has to be constructed by STAR for TE analysis

Question

[lkianmehr]

Hello,

I want to make index files before map RNA-seq reads on STAR. then I need to use theses alignments as an output of TEToolkit analysis. so two annotation files mask repeated and genes are provided. for Indexing as TEToolkit manual mentioned In order to map RNA-seq reads, STAR needs an index file of the reference genome and transcriptome. I made an index by using primary assemble reference genome (fasta file) and repeated mask gtf file, then mapped reads on. is it correct? because I just wanted to quantify TE elements, but TEtoolkit needs another option GTF file for gene, I think its index also should be provided. would appreciate any help?

Answer 1

[h.mon]

If your interest is TEs alone, you may index the genome with the repeats gtf. If you want to quantify both genes and TEs, concatenate the genes gtf and repeats gtf, and index the genome with this concatenated annotation - be sure to check if there are no common names between the TEs and genes gtfs.

[lkianmehr]

I did TE analysis by TEtranscript for only TE elements not genes, as you mentioned but there is no differentiation between two treated and control samples. however, I am interested more to compare the relative abundance of TE elements together, meanwhile, they even didn't show any significant difference with DESeq2. may I use their counts only to do something like that, if so how, do you have any suggestion in this case? I would appreciate if let me know