Skip to main content
Fig. 1 | BMC Bioinformatics

Fig. 1

From: Model selection and robust inference of mutational signatures using Negative Binomial non-negative matrix factorization

Fig. 1

Results from AIC, BIC, and SigMoS based on Po-NMF and NBN-NMF using simulated data. Each method is applied on different simulated data sets for four different types of noise: Poisson and Negative Binomial with dispersion parameter \(\alpha = 10, 200\) and \(\alpha \sim U(10,500)\). a The proportion of simulation runs where the number of signatures is correctly estimated. The true number of signatures varies in \(\{5,10\}\) and the number of patients in \(\{20,100,300\}\). The rectangular boxes highlight the results shown in b. The results are based on 100 simulation runs for scenarios with 20 and 100 patients and on 20 simulation runs for scenarios with 300 patients. b The estimated number of signatures in the range from 2 to 20 for 100 patients, where the true number of signatures is five

Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us