ANN-Spec v1.0

ANN-Spec was developed for finding un-gapped patterns in un-aligned DNA sequences. The program is primarily intended for finding transcription factor binding sites in promoter regions though it can be applied to protein sequences for the detection of conserved motifs as well.
ANN-Spec uses a neural network to spot regions of a given size (patterns) whose appearance in a set of sequences (the positive set) departs from its statistical expectance as determined analytically or after training with a background set (the negative set).
Please, paste or upload the positive set This is the set of sequences in FASTA format thought to contain the pattern


Please, paste or upload the negative set if you have one (or leave this blank if none) This is a set of sequences in FASTA format thought not to contain the pattern


If you don't have a training set, do you want to have a random training set automatically created for you?
Sequence type:
If yours are DNA sequences, do you want to use both strands (sense and complement) for training?
Number of sites expected from each sequence of the positive set
Width of the pattern to be learnt
Partition function: Defines the initial statistical distribution (expectancy) of the putative patterns being tested: You may choose
  • Analytical partition (if you have no background data, i.e. no negative set)
  • Random sites (to build an estimate from randomly sampled sites in the training set: this is specially useful if the training set is large enough and may contain the seeked pattern)
  • All sites (use all sites in the training data set, assumes that the pattern you seek is not present in the training set)
For publication of results, please cite:
Valverde, J. R. (2007) FInding motifs with ANN-Spec. embnet.news, vol. 13, no. 2, 13-18.
Workman, C. and Stormo, G.D. (2000) ANN-Spec: A method for discovering transcription factor binding sites with improved specificity. Proc. Pacific Symposium on Biocomputing 2000.
Heumann,J.M., Lapedes,A.S. and Stormo,G.D., Neural networks for determining protein specificity and multiple alignment of binding sites. Proccedings for the Intelligent Systems for Molecular Biology (ISMB), 1994;2:188-194. PMID: 7584389; UI: 96039019.
© José R. Valverde (www user interface) You can download this tool and install it at your site