Poly(A) Signal Miner is used to predict poly(A) signal in
human DNA sequences.
It was trained on
2327 terminal sequences including 1632 "unique" and
695 "strong" poly(A) sites. It was first used by
Erpin.
Our training accuracy is 78.16%
at 84.10% sensitivity and 71.54% specificity. Please refer to
our paper
and technology background and data
for more information about the classification model as well as
the training and validation data.
Restrictions:
Currently at most 1 sequence and 50,000 nucleotides per submission.