Biology subjects

Cicconet, M.

Publications and source records attributed to Cicconet, M..

9 recordsLinked to original sources

bioRxiv2021

Identification of two pathways mediating protein targeting from ER to lipid droplets

Pathways localizing proteins to their sites of action within a cell are essential for eukaryotic cell organization and function. Although mechanisms of protein targeting to many organelles have been defined, little is known about how proteins, such as key metabolic enzymes, target from the ER to cellular lipid droplets (LDs). Here, we identify two distinct pathways for ER-to-LD (ERTOLD) protein targeting: early ERTOLD, occurring during LD formation, and late ERTOLD, targeting mature LDs after their formation. By using systematic, unbiased approaches, we identified specific membrane-fusion machinery, including regulators, a tether, and SNARE proteins, that are required for late ERTOLD targeting. Components of this fusion machinery localize to LD-ER interfaces and appear to be organized at ER exit sites (ERES) to generate ER-LD membrane bridges. We also identified multiple cargoes for early and late ERTOLD. Collectively, our data provide a new model for how proteins target LDs from the ER. HIGHLIGHTSO_LIProteins localize to LDs either during formation or later through ER-LD bridges C_LIO_LISpecific membrane fusion machinery localizes to LDs and mediates protein targeting C_LIO_LISpecific ER exit site proteins associate with LDs and participate in ERTOLD targeting C_LIO_LIProteomic studies reveal cargoes for early and late ERTOLD targeting C_LI

cell biology↗

bioRxiv2021

UnMICST: Deep learning with real augmentation for robust segmentation of highly multiplexed images of human tissues

Newly developed technologies have made it feasible to routinely collect highly multiplexed (20-60 channel) images at subcellular resolution from human tissues for research and diagnostic purposes. Extracting single cell data from such images requires efficient and accurate image segmentation, a challenging problem that has recently benefited from the use of deep learning. In this paper, we demonstrate two approaches to improving tissue segmentation that are applicable to multiple deep learning frameworks. The first uses "real augmentations" that comprise defocused and saturated image data collected on the same instruments as the actual data; using real augmentation improves model accuracy to a significantly greater degree than computational augmentation (Gaussian blurring). The second involves imaging the nuclear envelope to better identify nuclear outlines. The two approaches cumulatively and substantially improve segmentation on a wide range of tissue types and provide a set of improved segmentation models. We speculate that the use of real augmentations may have applications in image processing outside of microscopy.

bioinformatics↗

bioRxiv2021

PuBliCiTy: Python Bioimage Computing Toolkit

The Python Bioimage Computing Toolkit (PuBliCiTy) is an evolving set of functions, scripts, and classes, written primarily in Python, to facilitate the analysis of biological images, of two or more dimensions, from electron or light microscopes. While the early development was guided by the goal of replacing an existing internal code-base with Python code, the effort later came to include novel tools, specially in the areas of machine learning infrastructure and model development. The toolkit is built on top of the so-called python data science stack, which includes numpy, scipy, scikit-image, scikit-learn, and pandas. It also contains some deep learning models, written in TensorFlow and PyTorch, and a web-app for image annotation, which uses Flask as the web framework. The main features of the toolkit are: (1) simplifying the interface of some routinely used functions from underlying libraries; (2) providing helpful tools for the analysis of large images; (3) providing a web interface for image annotation, which can be used remotely and on tablets with pencils; (4) providing machine learning model implementations that are easy to read, train, and deploy - written in a way that minimizes complexity for users without a computer science or software development background. The source code is released under an MIT-like license at github.com/hms-idac/PuBliCiTy. Details, tutorials, and up-to-date documentation can be found at the projects page as well. Project pagegithub.com/hms-idac/PuBliCiTy

bioinformatics↗

bioRxiv2019

Riffle Shuffle: A Supervised, Symmetry-Driven, GUI Toolkit for Mouse Brain Stack Registration and Plane Assignment

Immunostaining of brain slices is a ubiquitous technique used throughout neuroscience for the purposes of understanding the anatomical and molecular characteristics of brain circuits. Yet the variety of distortions introduced, and the manual nature of the preparation, hinder the use of the generated images from being rigorously quantified; instead most registration of brain slices is done laboriously by hand. Existing automated registration methods rarely make use of geometric shape information. When registering anterior-posterior brain slices, for example, small errors between consecutive planes accumulate, causing the symmetry axis of a plane to drift away from its starting position as depth increases. Furthermore, planes with imaging artifacts - e.g. one half of the slice is missing - can cause large errors, which are difficult to fix by changing global parameters. In this work we describe a method in which we register a set of consecutive brain slices enforcing all slices to have a vertical axis of symmetry, and then pair these slices optimally to planes from the Allen Mouse Brain Atlas via Dynamic Programming. The pipeline offers multiple human-in-the-loop opportunities, allowing users to fix algorithmic errors in various stages, including symmetry detection and pairwise assignment, via custom graphical interfaces. This pipeline enables large-scale analysis of brain slices, allowing this common technique to be used to generate quantitative datasets.

bioinformatics↗

bioRxiv2019

Molecular and anatomical organization of the dorsal raphe nucleus

The dorsal raphe nucleus (DRN) is an important source of neuromodulators in the brain and has been implicated in a wide variety of behavioral and neurological disorders. Although mostly studied as a source of serotonin, the DRN is comprised of multiple cell types that are subdivided into distinct anatomical subregions. However, the complex and incompletely characterized cellular organization of the DRN has impeded efforts to investigate the distinct circuit and behavioral functions of its subdomains. Here we used high-throughput single-cell RNA sequencing with in situ hybridization and viral tracing to develop a map of transcriptional and spatial profiles of cells in and around the mouse DRN. Our studies reveal the molecular and spatial organization of multiple neuron subtypes that are the cellular bases of functionally and anatomically distinct serotonergic subsystems, and provide a resource for the design of strategies for further dissection of these subsystems and their diverse functions.

neuroscience↗

bioRxiv2018

Highly multiplexed in situ protein imaging with signal amplification by Immuno-SABER

Probing the molecular organization of tissues requires in situ analysis by microscopy. However current limitations in multiplexing, sensitivity, and throughput collectively constitute a major barrier for comprehensive single-cell profiling of proteins. Here, we report Immunostaining with Signal Amplification By Exchange Reaction (Immuno-SABER), a rapid, highly multiplexed signal amplification method that simultaneously tackles these key challenges. Immuno-SABER utilizes DNA-barcoded antibodies and provides a method for highly multiplexed signal amplification via modular orthogonal DNA concatemers generated by Primer Exchange Reaction. This approach offers the capability to preprogram and control the amplification level independently for multiple targets without in situ enzymatic reactions, and the intrinsic scalability to rapidly amplify and image a large number of protein targets. We validated our approach in diverse sample types including cultured cells, cryosections, FFPE sections, and whole mount tissues. We demonstrated independently tunable 5-180-fold amplification for multiple targets, covering the full signal range conventionally achieved by secondary antibodies to tyramide signal amplification, as well as simultaneous signal amplification for 10 different proteins using standard equipment and workflow. We further combined Immuno-SABER with Expansion Microscopy to enable rapid and highly multiplexed super-resolution tissue imaging. Overall, Immuno-SABER presents an effective and accessible platform for rapid, multiplexed imaging of proteins across scales with high sensitivity.

cell biology↗

bioRxiv2018

FIT2 is a lipid phosphate phosphatase crucial for endoplasmic reticulum homeostasis

The endoplasmic reticulum (ER) protein Fat-Induced Transcript 2 (FIT2) has emerged as a key factor in lipid droplet (LD) formation, although its molecular function is unknown. Highlighting its importance, FIT2 orthologs are essential in worms and mice, and FIT2 deficiency causes a deafness/dystonia syndrome in humans. Here we show that FIT2 is a lipid phosphate phosphatase (LPP) enzyme that is required for maintaining the normal structure of the ER. Recombinant FIT2 exhibits LPP activity in vitro and loss of this activity in cells leads to ER membrane morphological changes and ER stress. Defects in LD formation in FIT2 depletion appear to be secondary to membrane lipid abnormalities, possibly due to alterations in phospholipids required for coating forming LDs. Our findings uncover an enzymatic role for FIT2 in ER lipid metabolism that is crucial for ER membrane homeostasis.

cell biology↗

bioRxiv2017

Bots for Software-Assisted Analysis of Image-Based Transcriptomics

We introduce software assistants - bots - for the task of analyzing image-based transcriptomic data. The key steps in this process are detecting nuclei, and counting associated puncta corresponding to labeled RNA. Our main release offers two algorithms for nuclei segmentation, and two for spot detection, to handle data of different complexities. For challenging nuclei segmentation cases, we enable the user to train a stacked Random Forest, which includes novel circularity features that leverage prior knowledge regarding nuclei shape for better instance segmentation. This machine learning model can be trained on a modern CPU-only computer, yet performs comparably with respect to a more hardware-demanding state-of-the-art deep learning approach, as demonstrated through experiments. While the primary motivation for the bots was image-based transcriptomics, we also demonstrate their applicability to the more general problem of scoring \"spots\" in nuclei.

bioinformatics↗

bioRxiv2017

Whole-Brain Serial-Section Electron Microscopy In Larval Zebrafish

Investigating the dense meshwork of wires and synapses that form neuronal circuits is possible with the high resolution of serial-section electron microscopy (ssEM)1. However, the imaging scale required to comprehensively reconstruct axons and dendrites is more than 10 orders of magnitude smaller than the spatial extents occupied by networks of interconnected neurons2--some of which span nearly the entire brain. The difficulties in generating and handling data for relatively large volumes at nanoscale resolution has thus restricted all studies in vertebrates to neuron fragments, thereby hindering investigations of complete circuits. These efforts were transformed by recent advances in computing, sample handling, and imaging techniques1, but examining entire brains at high resolution remains a challenge. Here we present ssEM data for a complete 5.5 days post-fertilisation larval zebrafish brain. Our approach utilizes multiple rounds of targeted imaging at different scales to reduce acquisition time and data management. The resulting dataset can be analysed to reconstruct neuronal processes, allowing us to, for example, survey all the myelinated axons (the projectome). Further, our reconstructions enabled us to investigate the precise projections of neurons and their contralateral counterparts. In particular, we observed that myelinated axons of reticulospinal and lateral line afferent neurons exhibit remarkable bilateral symmetry. Additionally, we found that fasciculated reticulospinal axons maintain the same neighbour relations throughout the extent of their projections. Furthermore, we use the dataset to set the stage for whole-brain comparisons of structure and function by co-registering functional reference atlases and in vivo two-photon fluorescence microscopy data from the same specimen. We provide the complete dataset and reconstructions as an open-access resource for neurobiologists and others interested in the ultrastructure of the larval zebrafish.

neuroscience↗