I am studying how a marine sponge holobiont reacts to particular environmental conditions (Differential gene expression in a treatment - control study). To do so, I sequenced 12 samples that contain the mixed totalRNA of the sponge (Eukaryote) and the 3 main symbionts (Prokaryotes), all of which I possess the assembled genome. To perform the alignment I need to consider the different organizations of their DNA. In particular, the fact that the sponge has splicing while the bacteria don't.
I am using Hisat2, which is based on a BWT mapping method.
Is it appropriate to perform 2 different alignments with different set-ups?
I explain: if I perform one alignment with splicing analysis and one without it and I use the first one only for the sponge and the second one only for the bacteria.
An alternative I have is to run everything performing the splicing analysis and remove the sequence considered spliced for the bacteria. However, the program that I am using doesn't seem to allow this filtering option.
Which would be the best option considering that I am performing a differential gene expression analysis on mixed samples between 2 different treatments? Would be correct to run 2 different analysis or the data would not be comparable?
Thank you
D.
Hi Davide Poli
Two suggestions;
First, You can classify the sequencing reads to each genome using hisat-2 or bowtie by using the commands --al (single-end), -al-conc (paired-end), and then perform the differential expression analysis (DE).
And second (and probably the most suitable in my opinion); perform a de novo assembly of all the reads, DE and then blastx to identify transcripts.
Best,
- Badges
- Science topic
- Similar topics
- Biological Science
- Genetics
- Gene Expression