[Bioc-sig-seq] question about trimLRPatterns

[Harris A. Jaffee]

On Sep 26, 2011, at 4:51 PM, wang peter wrote:
> dear harris:
> thank you very much for your previous help, but i am still confused  
> by such problems:
> [1] why does the second section of coding can not work,but the  
> first can
> subject = "TTTACGT"
> Lpattern = "TTTAACGT"
> trimLRPatterns(Lpattern = Lpattern, subject = subject,  
> max.Lmismatch=1,with.Lindels=TRUE)
> subject = "TGCATTT"
> Rpattern = "TGCAATTT"
> trimLRPatterns(Rpattern = Rpattern, subject = subject,  
> max.Rmismatch=1,with.Rindels=TRUE)
> Error in solveUserSEW(width(x), start = start, end = end, width =  
> width) :
>   solving row 1: 'allow.nonnarrowing' is FALSE and the supplied  
> start (0) is < 1

Let's hope this is the bug Herve mentioned yesterday, whose fix  
should appear soon.

> [2] how can i see the code of such functions:  
> which.isMatchingStartingAt, rev, normargPattern
> which are called by Biostrings:::.computeTrimEnd
> showMethods("which.isMatchingStartingAt") can not work

 > showMethods(which.isMatchingStartingAt, includeDefs=TRUE)
then
 > Biostrings:::.matchPatternAt

rev is in base (see ?rev)  [different from *reverse*, from IRanges]

 > Biostrings:::normargPattern

> [3] max.Rmismatch=0.1 will be replaced by 0.1*nchar(Rpattern) and  
> then by
> (-1,-1,...as.integer(0.1*nchar(Rpattern)))
> to better control each try, can i use max.Rmismatch=0.1*(1:nchar 
> (Rpattern))

Again, I can't tell what you're asking here.

> good luck
> shan gao