1
Define your query set
2
Choose platform and databases
Run it on the command line
Processing a lot of data? The same test runs locally with kycg, with no size limits, many queries at once.
# install kycg and yame
conda install -c zhou-lab -c conda-forge kycg yame
# fetch the knowledgebases for your array
kycg fetch -f EPIC
# a probe ID list, one ID per line
kycg test -p EPIC -m EPIC:CGI,ChromHMM,HM hits.txt
# a binary .cg query: pack one 0/1 per probe, in the array's probe order
yame pack -f b hits01.txt hits.cg
kycg test -m EPIC:CGI,ChromHMM,HM hits.cg
# lots of data: many queries in one run, one results table
kycg test -m EPIC:CGI,ChromHMM,HM -o results.tsv samples.cg more.cg
# universe restriction: -u keeps only the rows set in a record; repeat to combine
# (cg probes here; any 0/1 .cg over the array's rows works, e.g. your own mask)
kycg test -m EPIC:CGI -u ~/.local/share/yame/EPIC/KYCG/ProbeType.cm:cg hits.cg
See the KnowYourCG docs for all options and the YAME docs for packing .cg files.
Probe annotation
Results
Overview
Enriched databases (FDR < 0.01) across all groups. Dot size shows log2(odds ratio).
By database group
Top databases in the selected group, ranked by significance.
More views
Genes
Metagene
Enrichment along a gene model, from the MetagenePC knowledgebase: upstream of the TSS, across the gene body (0–100%), and downstream of the TES. Filled dots have FDR < 0.05.
Include the MetagenePC group in step 2 to see the metagene plot.