SCOPE++: A tool for homopolymer identification

[noms]

We are in the process of releasing a software tool SCOPE++ (Sequence Classification Of homoPolymer Emissions).

It has the capability to identify homopolymer sequences using Hidden Markov Models, allowing for fast, accurate and robust identification.

Additionally, poly(A)/poly(T) trimming is supported.

The executable can be found at the following website:

http://code.google.com/p/scopeplusplus/

[flxlex]

Could you explain why you choose to use HMM to detect homopolymer stretches? Why not use some simple perl code such as

$seq=uc($seq);
while ($seq=~/(A+|C+|G+|T+)/g){
	$homopolymer_length = length($1);
	...
}

Or is your code meant to detect imperfect long homoloymers?

[KevinLam]

hmm double posting here

http://seqanswers.com/forums/showthread.php?t=19156

to TS I think it's more helpful to have all comments on a single thread yeah?

[noms]

@flxlex:

Yup. It is used for detecting imperfect homopolymers. This is particularly useful when identifying poly(A) tails. Since base miscalls tend to become more frequent towards the ends of sequence reads, you aren't expected to identify perfect poly(A) tails.

@KevinLam:

Yeah. My bad. I'm linking these two threads.