Filtering Projection Images In Cryo-EM Single Particle Reconstruction

An important problem in the single-particle 3D reconstruction by cryo-electron microscopy(cryo-EM)is to construct the initial model of a macromolecule from its 2D noisy projection images at unknown random orientations.The first step of the 3D reconstruction is computing the directions of the projection images.The 'Angular construction' method estimates the directions of projections by establishing a coordinate system.However,it is difficult to obtain the projection angles of the projection images which have low signal-to-noise ratio.In this paper we propose a new method based on common lines to improve the initial model construction by filtering out low quality projection images.The proposed method improves the initial model by filtering out low quality projection images which we consider with high noise or wrong features.The projection angles are usually represented by Euler angles?,? and ?.It is found that the quality of a projection image can be evaluated in the process of estimating its Euler angle ?.After the low quality projection images are removed,the rest of the projection images are used to construct the initial model in our method.According to the rotation theorem,we use the Euler angles?,?,and ? of one projection image to calculate the angle of another projection image,then uses the consistency of to estimate the quality of the projection images.Finally,a part of low quality projection images are filtered out before constructing the initial model.At the same time,It is found that the common lines between the projection images also can be used to estimate the quality of the images.The more the good common lines a projection image has with the other projections,the higher quality the projection image is considered to have.We also found that filtering using Euler angle ? estimation is better than filtering using good common line estimation in the initial model construction.Although the common line filtering method can be used to select the high quality projection images,it is found that filtering using the Euler angle ? can produce better results.In the experiment,we simulate the projection images for 70 S E.coli ribosome.And we filter out a part of projection images by using the common line filtering algorithm and the Euler angle filtering algorithm.The experimental results show that both common line and Euler angle filtering algorithms can improve the accuracy of the initial model on the basis of the original ones,and Euler angle filtering algorithm is more effective than common line filtering algorithm.