Objective resolution measurement in single particle reconstructions based on a new spectral signal-to-noise ratio estimation

The resolution measurement of 3D reconstructed density map in single particle reconstruction is an important and still an open problem. In this paper, we propose a new protocol to measure the resolution just from the reconstructed density map. This approach estimates spectral signal-to-noise ratio (SSNR) of 3D reconstructed map by computing the ratio of signal power to noise power in frequency domain. The power distributions of signal and noise are estimated from structure particle region and surrounding region segmented by applying a mask procedure. The proposed protocol of calculating SSNR, which we term mask-SSNR (mSSNR), is independent of the reconstruction algorithms and can be used for density maps reconstructed with any reconstruction algorithm. Moreover, the mSSNR neither needs to split the dataset into halves like the Fourier shell correlation (FSC) approach, nor any original images or intermediate data like other SSNR calculation methods in this field. The mSSNR provides a direct calculation of SSNR based on its original definition, and is proven to be a better approach.