Skip to main content
. Author manuscript; available in PMC: 2023 Jul 21.
Published in final edited form as: Nat Mach Intell. 2022 Nov 15;4(11):1017–1028. doi: 10.1038/s42256-022-00561-w

Fig. 4 |. Evaluating gMVP and published methods in distinguishing rare DNMs in cases with neurodevelopmental disorders from those in controls.

Fig. 4 |

a, Distributions of gMVP-predicted scores for rare DNMs from ASD and NDD cases, as well as controls. We used a two-sided Mann–Whitney U test to assess the statistical significance of the difference between the cases and controls. Controls are unaffected siblings from the ASD study. b, Comparisons between gMVP and other published methods using DNMs from ASD cases and controls by precision–recall–proxy curves. Numbers on each point indicate rank percentile thresholds. The positions of the All Mis points are estimated from all missense variants without using any prediction method. c, The same comparison using data from NDD cases and controls.