Fernando Amat

Cortical representations of olfactory input by trans-synaptic tracing

In the mouse, each class of olfactory receptor neurons expressing a given odorant receptor has convergent axonal projections to two specific glomeruli in the olfactory bulb, thereby creating an odour map. However, it is unclear how this map is represented in the olfactory cortex. Here we combine rabies-virus-dependent retrograde mono-trans-synaptic labelling with genetics to control the location, number and type of ‘starter’ cortical neurons, from which we trace their presynaptic neurons. We find that individual cortical neurons receive input from multiple mitral cells representing broadly distributed glomeruli. Different cortical areas represent the olfactory bulb input differently. For example, the cortical amygdala preferentially receives dorsal olfactory bulb input, whereas the piriform cortex samples the whole olfactory bulb without obvious bias. These differences probably reflect different functions of these cortical areas in mediating innate odour preference or associative memory. The trans-synaptic labelling method described here should be widely applicable to mapping connections throughout the mouse nervous system.

Semantic-based indexing of fetal anatomies from 3-D ultrasound data using global/semi-local context and sequential sampling

The use of 3-D ultrasound data has several advantages over 2-D ultrasound for fetal biometric measurements, such as considerable decrease in the examination time, possibility of post-exam data processing by experts and the ability to produce 2-D views of the fetal anatomies in orientations that cannot be seen in common 2-D ultrasound exams. However, the search for standardized planes and the precise localization of fetal anatomies in ultrasound volumes are hard and time consuming processes even for expert physicians and sonographers. The relative low resolution in ultrasound volumes, small size of fetus anatomies and inter-volume position, orientation and size variability make this localization problem even more challenging. In order to make the plane search and fetal anatomy localization problems completely automatic, we introduce a novel principled probabilistic model that combines discriminative and generative classifiers with contextual information and sequential sampling. We implement a system based on this model, where the user queries consist of semantic keywords that represent anatomical structures of interest. After queried, the system automatically displays standardized planes and produces biometric measurements of the fetal anatomies. Experimental results on a held-out test set show that the automatic measurements are within the inter-user variability of expert users. It resolves for position, orientation and size of three different anatomies in less than 10 seconds in a dual-core computer running at 1.7 GHz.

Alphabet SOUP: A framework for approximate energy minimization

Many problems in computer vision can be modeled using conditional Markov random fields (CRF). Since finding the maximum a posteriori (MAP) solution in such models is NP-hard, much attention in recent years has been placed on finding good approximate solutions. In particular, graph-cut based algorithms, such as a-expansion, are tremendously successful at solving problems with regular potentials. However, for arbitrary energy functions, message passing algorithms, such as max-product belief propagation, are still the only resort. In this paper we describe a general framework for finding approximate MAP solutions of arbitrary energy functions. Our algorithm (called Alphabet SOUP for Sequential Optimization for Unrestricted Potentials) performs a search over variable assignments by iteratively solving subproblems over a reduced state-space. We provide a theoretical guarantee on the quality of the solution when the inner loop of our algorithm is solved exactly. We show that this approach greatly improves the efficiency of inference and achieves lower energy solutions for a broad range of vision problems.

Subtomogram alignment by adaptive Fourier coefficient thresholding

In the past few years, three-dimensional (3D) subtomogram alignment has become an important tool in cryo-electron tomography (CET). This technique allows one to produce higher resolution images of structures which can not be reconstructed using single-particle methods. Building on previous work, we present a new dissimilarity measure between subtomograms that works well for the noisy images that often occur in CET images. A technique that is more robust to noise provides the ability to analyze macromolecules in thicker samples such as whole cells or lower the defocus in thinner samples to push the first zero of the Contrast Transfer Function (CTF). Our method, Threshold Constrained Cross-Correlation (TCCC), uses statistics of the noise to automatically select only a small percentage of the Fourier coefficients to compute the cross-correlation, which has two main advantages: first, it reduces the influence of the noise by looking at only those peaks dominated by signal; and second, it avoids the missing wedge normalization problem since we consider the same number of coefficients for all possible pairs of subtomograms. We present results with synthetic and real data to compare our approach with other existing methods under different SNR and missing wedge conditions, and show that TCCC improves alignment results for datasets with SNR<0.1. We have made our source code freely available for the community.

Analysis of the Intact Surface Layer of Caulobacter crescentus by Cryo-Electron Tomography

The surface layers (S layers) of those bacteria and archaea that elaborate these crystalline structures have been studied for 40 years. However, most structural analysis has been based on electron microscopy of negatively stained S-layer fragments separated from cells, which can introduce staining artifacts and allow rearrangement of structures prone to self-assemble. We present a quantitative analysis of the structure and organization of the S layer on intact growing cells of the Gram-negative bacterium Caulobacter crescentus using cryo-electron tomography (CET) and statistical image processing. Instead of the expected long-range order, we observed different regions with hexagonally organized subunits exhibiting short-range order and a broad distribution of periodicities. Also, areas of stacked double layers were found, and these increased in extent when the S-layer protein (RsaA) expression level was elevated by addition of multiple rsaA copies. Finally, we combined high-resolution amino acid residue-specific Nanogold labeling and subtomogram averaging of CET volumes to improve our understanding of the correlation between the linear protein sequence and the structure at the 2-nm level of resolution that is presently available. The results support the view that the U-shaped RsaA monomer predicted from negative-stain tomography proceeds from the N terminus at one vertex, corresponding to the axis of 3-fold symmetry, to the C terminus at the opposite vertex, which forms the prominent 6-fold symmetry axis. Such information will help future efforts to analyze subunit interactions and guide selection of internal sites for display of heterologous protein segments.

Alignment of Cryo-Electron Tomography Datasets

Data acquisition of cryo-electron tomography (CET) samples described in previous chapters involves relatively imprecise mechanical motions: the tilt series has shifts, rotations, and several other distortions between projections. Alignment is the procedure of correcting for these effects in each image and requires the estimation of a projection model that describes how points from the sample in three-dimensions are projected to generate two-dimensional images. This estimation is enabled by finding corresponding common features between images. This chapter reviews several software packages that perform alignment and reconstruction tasks completely automatically (or with minimal user intervention) in two main scenarios: using gold fiducial markers as high contrast features or using relevant biological structures present in the image (marker-free). In particular, we emphasize the key decision points in the process that users should focus on in order to obtain high-resolution reconstructions.

3D segmentation of cell boundaries from whole cell cryogenic electron tomography volumes

Cryogenic electron tomography (cryo-ET) has gained increasing interest in recent years due to its ability to image whole cells and subcellular structures in 3D at nanometer resolution in their native environment. However, due to dose restrictions and the inability to acquire high tilt angle images, the reconstructed volumes are noisy and have missing information. Thus, features are unreliable, and precision extraction of the cell boundary is difficult, manual and time intensive. This paper presents an efficient recursive algorithm called BLASTED (Boundary Localization using Adaptive Shape and Texture Discovery) to automatically extract the cell boundary using a conditional random field (CRF) framework in which boundary points and shape are jointly inferred. The algorithm learns the texture of the boundary region progressively, and uses a global shape model and shape-dependent features to propose candidate boundary points on a slice of the membrane. It then updates the shape of that slice by accepting the appropriate candidate points using local spatial clustering, the global shape model, and trained boosted texture classifiers. The BLASTED algorithm segmented the cell membrane over an average of 93% of the length of the cell in 19 difficult cryo-ET datasets.