Fig. 1
From: Machine learning and statistics shape a novel path in archaeal promoter annotation
Overview of the classification rationale employed in this study. The figure is divided into i, ii, iii (train) and iv (test). (i) represents the conversion of genetic information into numeric attributes related to DDS, which is used as input of two classification methods. (ii) matches the Artificial Neural Network phase of classification; (iii) conveys information of how the classification was achieved through statistics. Both ii and iii were performed with experimentally verified promoters. Finally, the test, (iv) represents the validation process with upstream sequences whose promoters have not been identified yet. Each sequence undergoes through i, ii, and iii; then, the final decision is computed whether the sequence is a promoter or not