Skip to main content

Table 1 Results of predicting G-run regions in 100 real G4 sequences verified experimentally in [20].

From: Using hidden Markov models to investigate G-quadruplex motifs in genomic sequences

HMM

TP

FP

FN

SEN

PPV

F

Time (s)

Model 1

1196

3

0

1.000

0.998

0.999

0.008

Model 2

1113

2

83

0.931

0.998

0.963

0.014

Model 3

886

1

310

0.741

0.999

0.851

0.017

Model 4

1080

1

116

0.903

0.999

0.949

0.025

  1. In this test, we evaluated predictive performance of four HMM-based models using 10-fold cross validation. Time shows the total execution time to run the Viterbi algorithm on a machine with Intel Core i7-3820 3.60GHz CPU and 3.00GB RAM. The best performance is shown in boldface.
Back to article page

BMC Genomics

ISSN: 1471-2164

Contact us