Author: Jonas Pfab; Dong Si
Title: DeepTracer: Predicting Backbone Atomic Structure from High Resolution Cryo-EM Density Maps of Protein Complexes Document date: 2020_2_13
ID: er6lz09f_17
Snippet: The calculation of the confidence score between two atoms takes two factors into consideration: The Euclidean distance between them, as well as the average confidence of voxels that lay in between the atoms in the backbone prediction. Let ( , , ) be the normal probability density function at with mean and standard deviation , /,0 the Euclidean distance between atoms a and b, and /,0 the average backbone confidence between atoms a and b. We define.....
Document: The calculation of the confidence score between two atoms takes two factors into consideration: The Euclidean distance between them, as well as the average confidence of voxels that lay in between the atoms in the backbone prediction. Let ( , , ) be the normal probability density function at with mean and standard deviation , /,0 the Euclidean distance between atoms a and b, and /,0 the average backbone confidence between atoms a and b. We define the confidence score as shown in equation 1. The constant 3.8 was chosen as the mean for the distances between two atoms as this corresponds to the average distance found in nature. The standard deviation parameters of 5 and 0.3 where selected based on test runs. They were intentionally chosen higher than in the distribution found in solved structures, in order to cut some slack for prediction inaccuracies.
Search related documents:
Co phrase search for related documents- average distance and euclidean distance: 1, 2, 3, 4, 5
- confidence score and euclidean distance: 1
Co phrase search for related documents, hyperlinks ordered by date