Image-based cell phenotyping with deep learning

A cell's phenotype is the culmination of several cellular processes through a complex network of molecular interactions that ultimately result in a unique morphological signature. Visual cell phenotyping is the characterization and quantification of these observable cellular traits in images. Recently, cellular phenotyping has undergone a massive overhaul in terms of scale, resolution, and throughput, which is attributable to advances across electronic, optical, and chemical technologies for imaging cells. Coupled with the rapid acceleration of deep learning–based computational tools, these advances have opened up new avenues for innovation across a wide variety of high-throughput cell biology applications. Here, we review applications wherein deep learning is powering the recognition, profiling, and prediction of visual phenotypes to answer important biological questions. As the complexity and scale of imaging assays increase, deep learning offers computational solutions to elucidate the details of previously unexplored cellular phenotypes.     

Natural product-like chemical space: search for chemical dissectors of macromolecular interactions

Macromolecular interactions (i.e. protein-protein or DNA/RNA-protein interactions) play important cellular roles, including cellular communication and programmed cell death. Small-molecule chemical probes are crucial for dissecting these highly organized interactions, for mapping their function at the molecular level and developing new therapeutics. The lack of ideal chemical probes required to understand macromolecular interactions is the missing link in the next step of dissecting such interactions. Unfortunately, the classical combinatorial-chemistry community has not successfully provided the required probes (i.e. natural product inspired chemical probes that are rich in stereochemical and three-dimensional structural diversity) to achieve these goals. The emerging area of diversity-oriented synthesis (DOS) is beginning to provide natural product-like chemical probes that may be useful in this arena.     

Laying the groundwork for growth: Cell–cell and cell–ECM interactions in cardiovascular development

Cardiac development is reliant upon the spatial and temporal regulation of both genetic and chemical signals. Central to the communication of these signals are direct interactions between cells and their surrounding environment. The extracellular matrix (ECM) plays an integral role in cell communication and tissue growth throughout development by providing both structural support and chemical signaling factors. The present review discusses elements of cell–cell and cell–ECM interactions involved in cardiogenesis, and how disruption of these interactions can result in numerous heart defects. Examining the relationships between cells and their immediate environment has implications for novel and existing therapeutic strategies to combating congenital disorders. Birth Defects Research (Part C) 90:1–7, 2010. © 2010 Wiley‐Liss, Inc.     

Drug Repositioning by Kernel-Based Integration of Molecular Structure,Molecular Activity,and Phenotype Data

Computational inference of novel therapeutic values for existing drugs, i.e., drug repositioning, offers the great prospect for faster and low-risk drug development. Previous researches have indicated that chemical structures, target proteins, and side-effects could provide rich information in drug similarity assessment and further disease similarity. However, each single data source is important in its own way and data integration holds the great promise to reposition drug more accurately. Here, we propose a new method for drug repositioning, PreDR (Predict Drug Repositioning), to integrate molecular structure, molecular activity, and phenotype data. Specifically, we characterize drug by profiling in chemical structure, target protein, and side-effects space, and define a kernel function to correlate drugs with diseases. Then we train a support vector machine (SVM) to computationally predict novel drug-disease interactions. PreDR is validated on a well-established drug-disease network with 1,933 interactions among 593 drugs and 313 diseases. By cross-validation, we find that chemical structure, drug target, and side-effects information are all predictive for drug-disease relationships. More experimentally observed drug-disease interactions can be revealed by integrating these three data sources. Comparison with existing methods demonstrates that PreDR is competitive both in accuracy and coverage. Follow-up database search and pathway analysis indicate that our new predictions are worthy of further experimental validation. Particularly several novel predictions are supported by clinical trials databases and this shows the significant prospects of PreDR in future drug treatment. In conclusion, our new method, PreDR, can serve as a useful tool in drug discovery to efficiently identify novel drug-disease interactions. In addition, our heterogeneous data integration framework can be applied to other problems.     

Atomic Force Microscopy: A Promising Tool for Deciphering the Pathogenic Mechanisms of Fungi in Cystic Fibrosis

During the past decades, atomic force microscopy (AFM) has emerged as a powerful tool in microbiology. Although most of the works concerned bacteria, AFM also permitted major breakthroughs in the understanding of physiology and pathogenic mechanisms of some fungal species associated with cystic fibrosis. Complementary to electron microscopies, AFM offers unprecedented insights to visualize the cell wall architecture and components through three-dimensional imaging with nanometer resolution and to follow their dynamic changes during cell growth and division or following the exposure to drugs and chemicals. Besides imaging, force spectroscopy with piconewton sensitivity provides a direct means to decipher the forces governing cell–cell and cell–substrate interactions, but also to quantify specific and non-specific interactions between cell surface components at the single-molecule level. This nanotool explores new ways for a better understanding of the structures and functions of the cell surface components and therefore may be useful to elucidate the role of these components in the host–pathogen interactions as well as in the complex interplay between bacteria and fungi in the lung microbiome.

     

Contact response of cells can mediate morphogenetic pattern formation

Theories of morphogenetic pattern formation have included Turing's chemical prepatterns, mechanochemical interactions, cell sorting, and other mechanisms involving guided motion or signalling of cells. Many of these theories presuppose long-range cellular communication or other controls such as chemical concentration fields. However, the possibility that direct interactions between cells can lead to order and structure has not been seriously investigated in mathematical models. In this paper we consider this possibility, with emphasis on cells that reorient and align with each other when they come into contact. We show that such contact responses can account for the formation of multicellular patterns called parallel arrays. These patterns typically occur in tissue cultures of fibroblasts, and consist of clusters of cells sharing a common axis of orientation. Using predictions of a mathematical model and computer simulations of cell motion and interactions we show that contact responses alone, in the absence of other global controls, can promote the formation of these patterns. We suggest other situations in which patterns may result from direct cellular communication. Previous theories of morphogenesis are briefly reviewed and compared with this proposed mechanism.     

Mechanisms of action of disinfectants

Disinfectants and biocides are a chemically diverse group of agents which are generally considered to exhibit poor selective toxicity. This should not be mistaken for poor target specificity, however, and much is now known concerning the damaging interactions which may arise between bacterial cell and disinfectant agent. Critical governing features of these interactions are the physicochemical characteristics of the chemical agent, cell morphology, and the physiological status of the microorganism. Antibacterial events include membrane disruption, macromolecule dysfunction, and metabolic inhibition; the consequential effect is determined by the relative contribution(s) of the target(s) to microbial cell survival and the possible initiation of self-destructive processes. Disinfection kinetics offer a measure to differentiate between physiochemical and chemical interactions.Increasingly demanding disinfectant applications require more sophisticated use of biocidal systems. Approaches include: agents in combination, whereby a knowledge of mechanism of action assists in designing optimal mixtures; intracellular biocide delivery, using cellular transport processes to overcome cellular barriers; and targeted donation of biocide from delivery systems, requiring an understanding of target reactivity.A knowledge of disinfection mechanisms provides a basis from which novel chemistries and synergistic combinations may be developed.     

Exploring protein interactions by interaction-induced folding of proteins from complementary peptide fragments

The study of protein--protein interactions is central to understanding the chemical machinery that makes up the living cell. Until recently, facile methods to study these processes in intact, living cells have not existed. Furthermore, the assignment of function to novel proteins relies on demonstrating interactions of these proteins with proteins of known function. This review describes an experimental strategy, devised to study protein--protein interactions in any intact living cells based on protein-fragment complementation assays. Applications to quantitative analysis of interactions, allosteric processes and cDNA library screening are discussed. Recently, the feasibility of employing this strategy in genome-wide biochemical pathway mapping efforts has been demonstrated.     

Addressing chemical pollution in biodiversity research

Climate change, biodiversity loss, and chemical pollution are planetary-scale emergencies requiring urgent mitigation actions. As these “triple crises” are deeply interlinked, they need to be tackled in an integrative manner. However, while climate change and biodiversity are often studied together, chemical pollution as a global change factor contributing to worldwide biodiversity loss has received much less attention in biodiversity research so far. Here, we review evidence showing that the multifaceted effects of anthropogenic chemicals in the environment are posing a growing threat to biodiversity and ecosystems. Therefore, failure to account for pollution effects may significantly undermine the success of biodiversity protection efforts. We argue that progress in understanding and counteracting the negative impact of chemical pollution on biodiversity requires collective efforts of scientists from different disciplines, including but not limited to ecology, ecotoxicology, and environmental chemistry. Importantly, recent developments in these fields have now enabled comprehensive studies that could efficiently address the manifold interactions between chemicals and ecosystems. Based on their experience with intricate studies of biodiversity, ecologists are well equipped to embrace the additional challenge of chemical complexity through interdisciplinary collaborations. This offers a unique opportunity to jointly advance a seminal frontier in pollution ecology and facilitate the development of innovative solutions for environmental protection.     

Cell transformation by chemical agents--a review and analysis of the literature. A report of the U.S. Environmental Protection Agency Gene-Tox Program

The literature on cell transformation by chemical carcinogens has been critically reviewed. This subject is highly relevant to carcinogenesis in vivo, because the phenotypic changes that are collectively referred to as cell transformation usually involve the acquisition of tumorigenicity on inoculation into suitable rodent hosts. The systems chosen for review fall into 3 categories: cell strains (cells with a limited lifespan); cell lines (cells with an unlimited lifespan); and oncogenic viral-chemical interactions involving cells (Fischer rat embryo cells expressing an endogenous retrovirus, mouse embryo cells expressing the AKR leukemia virus, chemical enhancement of a simian adenovirus, SA7 transformation of Syrian hamster or rat embryo cells). Of the entire literature reviewed, 117 papers have been accepted for data abstraction by pre-defined criteria; these include 41 references to cell strains, 40 in cell lines, and 38 in viral-chemical interactions including cells. Because different systems have been reviewed, it would be meaningless to group all the compounds. The overall summary of the systems is as follows (many compounds have been tested in more than one system and, hence, are duplicated in these totals). (Chart: see text) In general, there is a reasonably good correlation between the results of the cell transformation systems and in vivo carcinogenesis. However, the many deficiencies of the EPA Merged Carcinogen List preclude definitive comparisons. Moreover, a number of 'false negatives' were obtained in systems that did not employ external metabolic activation. Further validation of all systems is required, but it seems very probable that several cell transformation systems will become valuable in assaying (with reasonable time and cost) the carcinogenic potential of environmental chemicals.     

A computational tool for identifying minimotifs in protein-protein interactions and improving the accuracy of minimotif predictions

Protein-protein interactions are important to understanding cell functions; however, our theoretical understanding is limited. There is a general discontinuity between the well-accepted physical and chemical forces that drive protein-protein interactions and the large collections of identified protein-protein interactions in various databases. Minimotifs are short functional peptide sequences that provide a basis to bridge this gap in knowledge. However, there is no systematic way to study minimotifs in the context of protein-protein interactions or vice versa. Here we have engineered a set of algorithms that can be used to identify minimotifs in known protein-protein interactions and implemented this for use by scientists in Minimotif Miner. By globally testing these algorithms on verified data and on 100 individual proteins as test cases, we demonstrate the utility of these new computation tools. This tool also can be used to reduce false-positive predictions in the discovery of novel minimotifs. The statistical significance of these algorithms is demonstrated by an ROC analysis (P = 0.001).     

FRET技术及其在蛋白质-蛋白质分子相互作用研究中的应用

简要综述了FRET方法在活细胞生理条件下研究蛋白质-蛋白质间相互作用方面的最新进展.蛋白质-蛋白质间相互作用在整个细胞生命过程中占有重要地位,由于细胞内各种组分极其复杂,因此一些传统研究蛋白质-蛋白质间相互作用的方法,例如酵母双杂交、免疫沉淀等可能会丢失某些重要的信息,无法正确地反映在当时活细胞生理条件下蛋白质-蛋白质间相互作用的动态变化过程.荧光共振能量转移（fluorescence resonance energy transfer, FRET）是近来发展的一项新技术,此项技术的应用,为在活细胞生理条件下对蛋白质-蛋白质间相互作用进行实时的动态研究,提供一个非常便利的条件.     

Genetic control of a structural polymer of the Neurospora crassa cell wall.

The heteropolysaccharide present in fraction 1 of the Neurospora crassa cell wall has been characterized in wild-type and morphological mutant strains of this fungus. Single and double mutations have been studied to determine possible genetic interactions controlling the chemical composition of such heteropolysaccharides . Single mutations studied were peak-2, scumbo ( FGSC 49), ragged ( FGSC 296), and crisp -1 ( FGSC 488). Double mutations studied were peak-2, scumbo ( FGSC 419), and ragged crisp -1. In all these strains, the main constituents of the heteropolysaccharide were glucose, mannose and galactose. Glycosidic linkages binding these neutral sugars have been identified by gas-liquid chromatography. A chemical structure of fraction I heteropolysaccharide is proposed. The results obtained with double mutants suggest the existence of genetic interactions, such as complementation or additive effects of lesions of different genes, to control the chemical composition and structure of the cell wall and the morphology of N. crassa mycelium.     

Platelet cancer cell interplay as a new therapeutic target

The interaction between circulating tumor cells and platelets is a key factor in cancer metastasis. These interactions, driven by a variety of receptors, support circulating tumor cells by protecting them from immune detection, cushioning them from shear stress, and promoting their arrest at the endothelium. Additionally, platelets have been shown to accumulate in the primary tumors, promoting tumor growth and angiogenesis by releasing growth factors. Furthermore, tumor cells can interact with platelets by inducing aggregation, which further protects cancer cells. However, the platelet cancer cell interplay also offers new approaches to develop targeted therapies. The accumulation of platelets in tumors has successfully been leveraged to deliver chemotherapeutics and imaging agents. Likewise, these platelet-based interactions have been utilized to target cancer cells in circulation. Although these current systems have limitations including drug loading and storage, leveraging platelet-cancer cell interactions to effectively target circulating tumor cells and tumors shows great promise for future cancer treatments.     

Probing the oligomeric state and interaction surfaces of Fukutin-I in dilauroylphosphatidylcholine bilayers

Fukutin-I is localised to the endoplasmic reticulum or Golgi apparatus within the cell, where it is believed to function as a glycosyltransferase. Its localisation within the cell is thought to to be mediated by the interaction of its N-terminal transmembrane domain with the lipid bilayers surrounding these compartments, each of which possesses a distinctive lipid composition. However, it remains unclear at the molecular level how the interaction between the transmembrane domains of this protein and the surrounding lipid bilayer drives its retention within these compartments. In this work, we employed chemical cross-linking and fluorescence resonance energy transfer measurements in conjunction with multiscale molecular dynamics simulations to determine the oligomeric state of the protein within dilauroylphosphatidylcholine bilayers to identify interactions between the transmembrane domains and to ascertain any role these interactions may play in protein localisation. Our studies reveal that the N-terminal transmembrane domain of Fukutin-I exists as dimer within dilauroylphosphatidylcholine bilayers and that this interaction is driven by interactions between a characteristic TXXSS motif. Furthermore residues close to the N-terminus that have previously been shown to play a key role in the clustering of lipids are shown to also play a major role in anchoring the protein in the membrane.     

Lipidomics of host-pathogen interactions

The cell biology of intracellular pathogens (viruses, bacteria, eukaryotic parasites) has provided us with molecular information of host-pathogen interactions. As a result it is becoming increasingly evident that lipids play important roles at various stages of host-pathogen interactions. They act in first line recognition and host cell signaling during pathogen docking, invasion and intracellular trafficking. Lipid metabolism is a housekeeping function in energy homeostasis and biomembrane synthesis during pathogen replication and persistence. Lipids of enormous chemical diversity play roles as immunomodulatory factors. Thus, novel biochemical analytics in combination with cell and molecular biology are a promising recipe for dissecting the roles of lipids in host-pathogen interactions.     

Coating of magnetic nanoparticles affects their interactions with model cell membranes

BackgroundThe use of functionalized iron oxide nanoparticles of various chemical properties and architectures offers a new promising direction in theranostic applications. The increasing applications of nanoparticles in medicine require that these engineered nanomaterials will contact human cells without damaging essential tissues. Thus, efficient delivery must be achieved, while minimizing cytotoxicity during passage through cell membranes to reach intracellular target compartments.MethodsDifferential Scanning Calorimetry (DSC), molecular modeling, and atomistic Molecular Dynamics (MD) simulations were performed for two magnetite nanoparticles coated with polyvinyl alcohol (PVA) and polyarabic acid (ARA) in order to assess their interactions with model DPPC membranes.ResultsDSC experiments showed that both nanoparticles interact strongly with DPPC lipid head groups, albeit to a different degree, which was further confirmed and quantified by MD simulations. The two systems were simulated, and dynamical and structural properties were monitored. A bimodal diffusion was observed for both nanoparticles, representing the diffusion in the water phase and in the proximity of the lipid bilayer. Nanoparticles did not enter the bilayer, but caused ordering of the head groups and reduced the area per lipid compared to the pure bilayer, with MAG-PVA interacting more strongly and being closer to the lipid bilayer.ConclusionsResults of DSC experiments and MD simulations were in excellent agreement. Our findings demonstrate that the external coating is a key factor that affects nanoparticle-membrane interactions. Magnetite nanoparticles coated with PVA and ARA did not destabilize the model membrane and can be considered promising platforms for biomedical applications.General significanceUnderstanding the physico-chemical interactions of different nanoparticle coatings in contact with model cell membranes is the first step for assessing toxic response and could lead to predictive models for estimating toxicity. DSC in combination with MD simulations is an effective strategy to assess physico-chemical interactions of coated nanoparticles with lipid bilayers.     

Toward in situ biochemistry: combining chemical kinetics approaches with biomolecular imaging in living cells

Our knowledge of protein-protein interactions comes primarily from experimentation with reconstituted proteins in dilute solutions. However, dilute solutions are poor approximations of the intracellular microenvironment, which contains exquisite and dynamic structure that is impossible to recreate inside test tubes. New approaches are needed that will allow the in situ characterization of protein-protein interactions inside living, intact cells. In this paper, we discuss recent efforts to measure the kinetics of protein binding within complexes inside living cells. While the experimental effort in these studies requires the confluence of techniques ranging from molecular imaging to cell and molecular biology, the experimental design and analysis requires a strong background in chemical kinetics and transport phenomena. Thus, we argue that chemical engineers can play a central role in furthering in situ approaches to cellular analysis. Such efforts may aid significantly in advancing quantitative knowledge of cellular signaling and physiology.     

Glycosaminoglycan interaction networks and databases

Glycosaminoglycans are complex polysaccharides exhibiting a large structural and conformational diversity. These key biological players organize the extracellular matrix, contribute to cell–matrix interactions, and regulate cell signaling. Natural and synthetic libraries of glycosaminoglycans have been spotted on microarrays to find glycosaminoglycan partners and determine the size and the chemical groups promoting protein binding. Advances in glycosaminoglycan sequencing allow the characterization of glycosaminoglycan sequences interacting with proteins, and glycosaminoglycan-mediated pull-down proteomics can identify glycosaminoglycan-binding proteins at a proteome scale in various biological samples. The analysis of the glycosaminoglycan interaction networks generated using these data gives insights into the molecular and cellular mechanisms underlying glycosaminoglycan functions. These interactomes can also be used to design inhibitors targeting specific GAG interactions for therapeutic purpose.