view article

Figure 4
LLG as a function of the calculated normalized intensity, ZC, for several levels of information gain. In each case the observed normalized intensity, ZO, is 2.0 and the value of σA is 0.8, which would correspond, for instance, to an r.m.s. error of 0.4 Å for a model at a resolution of 2.2 Å. The LLG computed with the formula for LLGI (Read & McCoy, 2016BB18) is shown in blue and the LLG computed with the inflated-variance Rice function LLGFW is shown in orange. (a) Information gain is 0.01 bits, corresponding to [\sigma_{Z_{\rm O}}] = 2.57, EFW = 0.91, [{\sigma _{{E_{\rm FW}}}}] = 0.44. (b) Information gain is 0.1 bits, corresponding to [\sigma_{Z_{\rm O}}] = 1.33, EFW = 0.99, [\sigma_{E_{\rm FW}}] = 0.41. (c) Information gain is 1 bit, corresponding to [\sigma_{Z_{\rm O}}] = 0.63, EFW = 1.25, [\sigma_{E_{\rm FW}}] = 0.26. (d) Information gain is 3 bits, corresponding to [\sigma_{Z_{\rm O}}] = 0.21, EFW = 1.40, [\sigma_{E_{\rm FW}}] = 0.08.

Journal logoSTRUCTURAL
BIOLOGY
ISSN: 2059-7983
Follow Acta Cryst. D
Sign up for e-alerts
Follow Acta Cryst. on Twitter
Follow us on facebook
Sign up for RSS feeds