saaientist.R

# A Possible solution to http://saaientist.blogspot.com/2010/06/encounter-with-incanter-about-clojure.html
# in R

data = data.frame( position=c( 123, 245, 832, 1234, 2345, 8315, 9213 ), 
                   qual=    c(  29,  93,  51,   63,   75,    9,   59 ),
                   ref=     c( 'A', 'C', 'C',  'T',  'A',  'C',  'T' ),
                   alt=     c( 'G', 'G', 'T',  'G',  'C',  'A',  'G' ) )

ti.data = data[ paste( data$ref, data$alt, sep='' ) %in% c( 'AG', 'CT' ), ]
tv.data = data[ paste( data$ref, data$alt, sep='' ) %in% c( 'AC', 'CG', 'GT', 'AT' ), ]

t( sapply( seq( from=10, to=100, by=10 ), function( i ) {
	ti.count = dim( ti.data[ ( ti.data$qual <= i & ti.data$qual > i - 10 ), ] )[1]
	tv.count = dim( tv.data[ ( tv.data$qual <= i & tv.data$qual > i - 10 ), ] )[1]
	list( cutoff=i, ratio=ti.count / tv.count )
} ) )

Don't know if this even works... Not enough input data ;-)