# Name

defocus_get - Estimate defocus from 1D rotationally averaged power spectrum. This function is semi-automated, and requires the user to give correct cut_off frequency to determine the area of power spectrum for defocus estimation.

# Usage

defocus = defocus_get((fnam_roo, volt=300,ps=1,cs=2,amplitude_contrast=.1,f_start=30,f_stop=8,docf="a",skip="#",round_off=100,nr1=3,nr2=6)

## Input

- fnam_roo
- document file name of 1D rotational average of power spectrum
- volt
- voltage
- ps
- pixel size of the micrograph
- cs
- Sherical abberation constant
- amplitude_constrast
- amplitude constrast ratio
- f_start
- low frequency limit (in Ansgtrom) of selecting power spectrum containing strong CTF for defocus estimation. The default value is 30 Angstrom.
- f_stop
- high frequency limit (in Angstrom) of selecting power spectrum containing strong CTF for defocus estimation. The default value is 8 Angstrom.
- docf
- 1D power spectrum document file format, either standard spider one "s" or arbitrary format "a"
- skip
- the mark of comments in the document file, such as ";", or "#"
- round_off
- round_off error,by default it is set as 100 Angstrom. By adjusting this number, the defocus estimation can be accurate to any digits the user expects.
- nr1
- the highest rank of polynomials used to fit envelop function
- nr2
- the highest rank of polynomials used to fit background noise

## Output

- defocus
- the estimated defocus in Angstrom.

## Options

# Description

- The function requires the user to define the area in the 1D rotationally averaged power spectrum for defocus estimation.
- Empirically power spectrum from 1/30 Ansgtrom to 1/8 Angstrom contains discernable CTF traces, and tends to give estimation which is close to the true defocus setting.
- The starting frequency is supposed to exclude the protein peak, and other spurious peaks caused by artifacts.
- The stopping freqeuncy is supposed to exclude too flat area of the power spectrum in high frequency area, which tends to underestimate the defocus.
- This function doesn't incude astigmatsm estimation.

# Method

- Constrained simplex method.

# Reference

.1. 1. Barrodale and F.D.K. Roberts L1 solution to linear equations subject to linear equality and inequality constraints ACM TOMS 6 (1980), 231-235.

.2. Z. Huang, P. R. Baldwin, S.Mullapudi, and P .A. Penczek, Automated determination of parameters describing power spectra of micrograph images in electron microscopy. J. Struct. Biol. 144 (2003), pp. 79–94.

# Author / Maintainer

Zhong Huang

# Keywords

- category 1
- UTILITIES

# Files

utilities.py

# See also

# Maturity

- stable
- works for most people, has been tested; test cases/examples available.

# Bugs

None. It is perfect.