FASTDEF: fast defocus and astigmatism estimation for high-throughput transmission electron microscopy

Area of Science:

• Microscopy and Imaging
• Computational Biology
• Materials Science

Background:

• Accurate contrast transfer function (CTF) estimation is crucial for high-resolution transmission electron microscopy (TEM).
• Existing CTF estimation methods can be time-consuming and may require initial defocus parameter estimations.
• High-throughput TEM imaging demands faster and more automated analysis techniques.

Purpose of the Study:

• To develop a fast, automated algorithm for CTF estimation in TEM.
• To improve the efficiency of CTF estimation for high-throughput applications.
• To provide accurate defocus and astigmatism estimations comparable to established methods.

Main Methods:

• The algorithm utilizes a quadrature filter to obtain the wrapped phase of the power spectra density pattern.
• Phase unwrapping is performed to generate a continuous phase map.
• Zernike polynomial fitting is applied to determine defocus and astigmatism parameters.

Main Results:

• The developed algorithm is significantly faster (nearly an order of magnitude) than existing approaches.
• It achieves accuracy in defocus and astigmatism estimation comparable to Xmipp and CTFFIND3.
• The method demonstrates successful CTF estimation across various sample types, including negative stain, cryo-EM (with and without carbon support), ice-embedded samples, and tilted specimens.

Conclusions:

• The new algorithm provides a fast, automated, and accurate solution for CTF estimation in TEM.
• Its suitability for high-throughput work is demonstrated, requiring no initial defocus estimation.
• The algorithm is extendable for higher-order aberration estimation and is available within the Xmipp package.