Motivation: a new approach to the prediction of eukaryotic Pol II promoters from DNA sequence takes advantage of a combination of elements similar to neural networks and genetic algorithms to recognize a set of discrete subpatterns with variable separation as one pattern, a promoter. The neural networks use as input a small window of DNA sequence, as well as the output of other neural networks. Through the use of genetic algorithms, the weights in the neural networks are optimized to maximally discriminate between promoters and non-promoters.

Results: after several thousand generations of optimization, the algorithm was able to discriminate between vertebrate promoter and non-promoter sequences in a test set with a correlation coefficient of 0.63. In addition, all five known transcription start sites on the plus strand of the complete Adenovirus genome were within 161 bp of 35 predicted transcription start sites. On standardized test sets consisting of human genomic DNA, the performance of Promoter 2.0 compares well with other software developed for the same purpose.