Co-expression analysis using Camoco
Camoco (Schaefer et al., 2018) was used to prioritize causal RSA-related genes in maize. The GWAS locus information was derived from the genome-wide association analysis, and density metrics were used to perform the co-expression network analysis. The candidate window size was 50 kb, and the maximum number of flanking genes was two. The candidate windows were calculated both upstream and downstream of the input SNPs.
#Building RefGen Object (ZmB73_5b_FGS.gff, this file can be downloaded from here (https://pan.baidu.com/s/1ZZpsXj41_9YsyamnQSZrew) using the extraction code (xaj9)).
$ camoco build-refgen
ZmB73_5b_FGS.gff
"Zm5bFGS"
"Maize 5b Filtered Gene Set"
5b
"Zea mays"
#Building COB object (co-expression network)
$ camoco build-cob
Root_FPKM_35lines.txt
ZmRoot_35lines
"Maize Root Network 19CA_HN35lines"
Zm5bFGS
#Building Ontology Datasets
$ camoco build-go
zm_go.tsv
go.obo
"ZmGO"
"Maize GO"
Zm5bFGS
#Building GWAS Object
$ camoco build-gwas
ZmRoot20_merge8traits.allLocs.csv
ZmRoot20CMLM_8traits
"Maize Root 0-20 GWAS"
Zm5bFGS
--sep ','
--trait-col
roottraits
--chrom-col chr
--pos-col pos
#Calculating co-expression
$ camoco overlap
ZmRoot_35lines
density
--gwas ZmRoot0CMLM_8traits
--term merge8traits
--candidate-window-size 50000
--candidate-flank-limit 2
Reference
Schaefer, R.J. et al. Integrating coexpression networks with gwas to prioritize causal genes in maize. Plant Cell 30, 2922-2942 (2018).