Journal of Cheminformatics

Table 7 Slopes of mean σ_ŷ and σ_ŷ 95% CI versus σ for the Gaussian Process algorithm. Results are shown with and without the input of σ_y into the algorithm

From: The effect of noise on the predictive limit of QSAR models

Dataset	No σ_y	No σ_y	With σ_y	With σ_y
Dataset	Mean σ_y	σ_y 95% CI	Mean σ_y	σ_y 95% CI
G_298_atom	1.0	0.40	0.52	− 0.10
Alpha	1.1	0.16	0.44^a	0.32^a
Solv	0.94	− 0.19	0.10	0.10
BACE	0.25	0.38	− 0.12	− 0.35
Tox_102	0.32	0.028	− 0.96	− 0.48
Tox_134	0.49	0.53	− 0.66	− 0.17
LD₅₀	0.66	− 0.39	− 0.60	0.14

^aThe first point was omitted in these calculations because of a discontinuity in the line

Back to article page

ISSN: 1758-2946

Contact us

Submission enquiries: journalsubmissions@springernature.com