search

laurahspencer / DuMOAR

0 stars 0 forks source link

New RNASeq approach: de-novo transcriptome from MEGAN-scrubbed reads, Salmon for counts #40

Closed laurahspencer closed 11 months ago

laurahspencer commented 1 year ago

@ggoetznoaa build a new transcriptome using the reads filtered to use only Arthropoda and Not Assigned reads (as per Megan), then generate counts using Salmon. Provide Laura with gene count matrix.

Laura to perform differential expression analysis.

ggoetznoaa commented 11 months ago

Ok, I built a new transcriptome using Trinity v2.14.0 and all the MEGAN6 filtered reads. It generated 191492 contigs and 142402 'genes'. I then mapped the MEGAN6 reads using salmon v1.9.0 to the new transcriptome. I was getting roughly the following mapping rate percent stats.

 Min.   :92.72  
 1st Qu.:95.40  
 Median :95.89  
 Mean   :95.53  
 3rd Qu.:96.02  
 Max.   :96.66
laurahspencer commented 11 months ago

@ggoetznoaa posted results to https://github.com/laurahspencer/DuMOAR/tree/main/results/salmon