2020
DOI: 10.1016/j.jsb.2020.107545
|View full text |Cite
|
Sign up to set email alerts
|

Mitigating local over-fitting during single particle reconstruction with SIDESPLITTER

Abstract: Graphical abstract

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

2
75
0

Year Published

2020
2020
2024
2024

Publication Types

Select...
9
1

Relationship

1
9

Authors

Journals

citations
Cited by 98 publications
(77 citation statements)
references
References 44 publications
2
75
0
Order By: Relevance
“…While this shares the motivation for non-uniform refinement, it relies on manual interaction during refinement, often necessitating a tedious trial and error process that can be difficult to replicate. SIDESPLITTER 31 is a recently proposed method to mitigate over-fitting during refinement. Based on estimates of signal-to-noise ratios from voxel statistics, it smooths local regions with low signal-to-noise more aggressively than indicated by a global FSC-based resolution.…”
Section: Discussionmentioning
confidence: 99%
“…While this shares the motivation for non-uniform refinement, it relies on manual interaction during refinement, often necessitating a tedious trial and error process that can be difficult to replicate. SIDESPLITTER 31 is a recently proposed method to mitigate over-fitting during refinement. Based on estimates of signal-to-noise ratios from voxel statistics, it smooths local regions with low signal-to-noise more aggressively than indicated by a global FSC-based resolution.…”
Section: Discussionmentioning
confidence: 99%
“…To aid de novo model building of cytoplasmic and luminal domains, these parts were subjected to masked focused classification (K = 5, T = 8), 3D autorefinement and post-processing in Relion, yielding improved maps at 3.4 Å and 3.2 Å, respectively. To obtain highest quality maps of the transmembrane domains, the 177560 particles from consensus refinement were processed using Sidesplitter ( Ramlaul et al, 2020 ), producing a 3.3 Å global map after Relion post-processing, where transmembrane helix pitch and side chains were well resolved and allowed for unambiguous sequence assignment. The final particle set was further subjected to 3D variability analysis ( Punjani and Fleet, 2020 ) in cryoSPARC, revealing the presence or absence of the EMC7 lumenal domain between the EMC1 beta-propellers.…”
Section: Methodsmentioning
confidence: 99%
“…In the final refinement sequence, half maps were locally filtered between refinement iterations using SIDESPLITTER 63 , an adaptation of the LAFTER algorithm 64 that maintains gold-standard separation between the two half maps. The final model contained 403,991 particles and reached an overall resolution of 3.3 Å with side chains visible for most of the complex (Extended Data Figs.…”
Section: Data Processing and Model Buildingmentioning
confidence: 99%