Pipeline for processing RNAseq data from Flaveria. Four species, two C3 and two C4.
Flaveria species - C3:
- Flaveria pringlei
- Flaveria robusta
Flaveria species - C4:
- Flaveria trinervia
- Flaveria bidentis
Data contains samples from six sections of leaves from each species with three replicates of each section.
Fastqc
- Quality checking raw reads
- Check for over-represented kmers/sequences.
Trimmomatic
- Removing adapters?
- Trimming low quality sequences at the ends of reads. Cutoff set to 10? 15? [1]
Error Correction
- Read error correction using hamming trees.
- Reptile initially chosen because it came out best in this paper [2]
- BayesHammer then used because Reptile has problems [5]
Khmer
- preparing reads for assembly
- reduce coverage
Assemblotron
- optimal transcriptome assembly
Annotation
- Reciprocal best usearch transcripts against Arabidopsis
- Round-robin rbusearch between flaveria species to update annotation
Expression quantification
- mapping reads output from trimmomatic to transcripts
- using Express[3] or Sailfish[4]
Differential Expression Analysis
- using EBSeq or possibly BaySeq
Type rake
[1] Paper about optimising trimmomatic cutoff
[2] Paper comparing read error correction methods
[3] Paper about Express
[4] Paper about Sailfish
[5] Paper about BayesHammer