Skip to main content
Fig. 1 | Plant Methods

Fig. 1

From: PredCRG: A computational method for recognition of plant circadian genes by employing support vector machine with Laplace kernel

Fig. 1

a Box plot of the sequence lengths of the positive dataset, where it can be seen that sequence length with more than 1000 amino acids are outlying observations. Thus, the maximum sequence length considered is 1000 amino acids. b Overall accuracy for the four homogeneous sub-datasets and the heterogeneous full dataset. It is seen that accuracies are higher for the sub-datasets with homogeneous sequence length as compared to dataset with highly heterogeneous sequence length. c Performance metrics for seven different kernel functions with respect to classification of circadian and non-circadian proteins using support vector machine. Among all the kernel functions, Laplace, linear and radial kernels are found to be superior with regard to overall classification accuracy

Back to article page