Grouping data.frame by column name pattern in R

This is kind of brutal force solution. First you transpose the data.frame Then use aggregate function Finally transpose back. An example...

x = data.frame(a1=rnorm(10), a2=rnorm(10), b1=rnorm(10), b2=rnorm(10))
g=sapply(colnames(x), function(xx) {strsplit(xx,split='')[[1]][1]})
y = t(aggregate(t(x),by=list(g),mean, simplify=T))
yy = data.frame(y[-1,])
colnames(yy) = y[1,]
rownames(yy) = rownames(x)