PIK3CA-MAPK_co-oc.Rnw

<<setup_tp53, include=FALSE>>=
io = import('io')
ar = import('array')
df = import('data_frame')
st = import('stats')
tcga = import('data/tcga')
util = import('./util_2')

# define the genes we're interested in
genes_mut = c("PIK3CA", "BRAF", "KRAS", "NRAS")
genes_cna = c("EGFR_amp", "KRAS_amp", "ERBB2_amp")

# load tcga scores for p53
mapk = io$load('../scores/tcga/pathways_mapped/speed_matrix.RData')[,'MAPK',drop=FALSE]

# load mutation status for the genes
path = '../analyses/tcga_mutation'
mut = io$load(file.path(path, 'mut_driver_matrix.RData'))[,genes_mut,drop=FALSE]
cna = io$load(file.path(path, 'cna_driver_matrix.RData'))[,genes_cna,drop=FALSE]
data = ar$stack(list(mut, cna), along=2)

ar$intersect(mapk, data, along=1)
study = tcga$barcode2study(rownames(data))

PIK3CA = data[,"PIK3CA"]
BRAF = data[,"BRAF"]
KRAS = data[,"KRAS"]
NRAS = data[,"NRAS"]
EGFR_amp = data[,"EGFR_amp"]
KRAS_amp = data[,"KRAS_amp"]
ERBB2_amp = data[,"ERBB2_amp"]


# all still significant after regressing out TP53 mods
assocs = st$lm(mapk ~ study + BRAF + KRAS + NRAS + EGFR_amp + KRAS_amp + ERBB2_amp + PIK3CA)
@