Three-dimensional deconvolution processing for STEM cryotomography

Abstract

The complex environment of biological cells and tissues has motivated development of three-dimensional (3D) imaging in both light and electron microscopies. To this end, one of the primary tools in fluorescence microscopy is that of computational deconvolution. Wide-field fluorescence images are often corrupted by haze due to out-of-focus light, i.e., to cross-talk between different object planes as represented in the 3D image. Using prior understanding of the image formation mechanism, it is possible to suppress the cross-talk and reassign the unfocused light to its proper source post facto. Electron tomography based on tilted projections also exhibits a cross-talk between distant planes due to the discrete angular sampling and limited tilt range. By use of a suitably synthesized 3D point spread function, we show here that deconvolution leads to similar improvements in volume data reconstructed from cryoscanning transmission electron tomography (CSTET), namely a dramatic in-plane noise reduction and improved representation of features in the axial dimension. Contrast enhancement is demonstrated first with colloidal gold particles and then in representative cryotomograms of intact cells. Deconvolution of CSTET data collected from the periphery of an intact nucleus revealed partially condensed, extended structures in interphase chromatin.

Entropy Regularized Deconvolution of Cellular Cryo-Transmission Electron Tomograms

“…However, in our experience, the DC process has never been observed to invent a structure that is not present in the WBP raw data control. (26). This study makes the statement that the missing wedge of information is substantially filled by DC.…”

“…However, in our experience, the DC process has never been observed to invent a structure that is not present in the WBP raw data control. (26).…”

“…Specifically, in cellular images, high intensities and high second-order derivatives exhibit certain sparse distribution, and this property is exploited by the custom regularization used in ER-DC. This regularization was originally designed for fluorescence images, and this approach was taken recently for processing of STEM cryo-tomography (CSTET) reconstructions (26). It is similar to deconvolution applied to fluorescence microscopy, where out of focus light creates a haze, but differs in that the artifacts to be removed originate primarily in the reconstruction rather than the optics.…”

“…Because of these issues with cryo-ET data, filters to improve contrast and compensate for the missing wedge are an area of ongoing research (18). These techniques include non-linear anisotropic diffusion (NAD), convolutional neural networks based on de-tector noise models, wavelet based filtering methods, different implementations of deconvolution, and model based iterative reconstruction (MBIR) (19)(20)(21)(22)(23)(24)(25)(26)(27)(28). Here, we present a deconvolution approach to achieve both enhanced SNR and missing wedge compensation.…”

A Proposed Unified Interphase Nucleus Chromosome Structure: Preliminary Preponderance of Evidence

“…A recent publication described deconvolution (DC) of STEM tomography of cryo-preserved cellular structures (3 and references therein). In brief, cells were plunge-frozen, preserving the aqueous cellular structures in a glassy ice medium, then micron-sized slabs were selected from thin regions of whole cells, followed by STEM tomography, and DC data processing (3). The DC substantially filled the missing wedges, resulting from incomplete tilts, regions that had in the past greatly compromised the Z resolution, allowing much greater ease in interpretation of the tomograms.…”