Pipeline for mRNA-seq analysis.

To be run after fastq files have been trimmed with cgat-flow readqc, runs standard mRNA-seq analysis and QC.

1. mapping
   • map reads with STAR and index BAM files
   • if mapping is conducted with CGAT pipeline, these BAMs can be used as input, rather than running STAR
2. readcounts
   • count reads over genes with featureCounts
3. summarystats
   • collect mapping statistics with picard tools
4. readquant
   • calculate TPMs with Salmon
5. coverage
   • generate bigWig coverage tracks for visualisation

• fastq.gz formatted files. Can be paired or single end.
• should be named sample_r1.fastq.[1-2].gz (PE) or sample_r1.fastq.gz (SE)
• naming convention: sample names should be informative e.g. "group_condition_treatment_replicate.fastq.1.gz" as they're used to generate a sample information table to annotate plots and create comparisons for DESeq2
• fastq files should be placed in a folder named "data.dir/"
• alternatively BAMs can be placed in "bam.dir/" and only tasks after mapping will be run

• bam.dir: mapped reads, bigWigs
• read_counts.dir: raw read counts
• salmon.dir: TPMs
• csvdb: sqlite3 db containing all QC metrics, raw read counts, TPMs, etc.
• Jupyter notebook reports: QC and DESeq2 analysis

• cgat-core, cgat-flow, cgat-apps (https://github.com/cgat-developers)
• ruffus 2.8.3
• jupyter-notebook 6.0.2
• rpy2 3.2.4
• pandas 0.25.3
• numpy 1.17.3
• pybedtools 0.8.0
• seaborn 0.9.0
• matplotlib 3.1.1
• sqlite3
• yaml
• glob
• os
• re

• Salmon 0.11.3
• STAR 2.5.1b
• picard tools 2.10.9
• samtools 1.9
• subread 1.6.3
• deeptools 3.7.3

• ggplot2
• gplots
• gridExtra
• scales
• ggrepel
• wesanderson
• reshape2
• dplyr
• plyr
• grid
• RColorBrewer
• ComplexHeatmap
• circlize
• dendextend
• Rtsne
• DESeq2
• vsn
• topGO
• stringr