Skip to main content
Figure 2 | BMC Bioinformatics

Figure 2

From: MiRPara: a SVM-based software tool for prediction of most probable microRNA coding regions in genome scale sequences

Figure 2

Parameters defined for miRNA prediction. A total of 77 parameters were initially selected that described properties of the pri-miRNA, pre-miRNA and miRNA sequences. These parameters were initially classified as Descriptive or related to the Size, Stability, Sequence or Structure properties. Parameters that were too broad in their range of values for use with an SVM were removed before training and parameter filtering was used to reduce the the remaining set to a subset of key parameters that were sufficient for accurate miRNA prediction. o - SVM parameters selected by the Overall models after parameter filtering, a - SVM parameters selected by the Animal model after parameter filtering, p - SVM parameters selected by Plant models after parameter filtering, Gray Background - SVM parameters selected by all three models after parameter filtering. See Materials and Methods for full details.

Back to article page