Skip to main content
. 2020 Jul 28;37(2):147–154. doi: 10.1093/bioinformatics/btaa672

Fig. 6.

Fig. 6.

Benchmarks on synthetic data. When the number of samples is small (M =3) or data complexity (i.e. the number of clusters) is low (K =2), the Flat model performs comparably to the GP0 models. In all other cases, it is outperformed by them. Parameters used in data simulation were informed by the experimental data (see main text for details)