Using a combination of classical microbiology and NGS, we studied the colistin resistant fraction of the BE microbiome in 20 Stockholm subway stations. This talk will highlight the methodology and partial results of a longitudinal dataset currently being assembled.
It is always difficult to predict exciting results for great paper. More and more rich labs are using screening strategy to get ideas. These screens are time-consuming and expensive way to discover something new. There is also another way to big discoveries but they require step-by-step development of the project based on small observation the experimenter had noticed. Within my talk I will present the story built on the interesting observed unexpected phenotype. I will go through the experiments made within time period of about 5 years to show the main milestones and decisions we have made. Even when all data was generated one additional experiment forced us to look on these data completely from unexpected side.
Hydrogen sulfide to the rescue: Trials and tribulations of translational research
Ulceration of the gastrointestinal tract by nonsteroidal anti-inflammatory drugs (NSAIDs) remains a major health problem. We have exploited the powerful protective and repair-promoting effects of hydrogen sulfide to design new NSAIDs that are safe in the GI tract. Translating our basic science research to clinical use has been exciting but very challenging.
Adhesion proteins more than just cardiomyocytes coupling
The structural integrity of the heart is necessary for its function and is maintained by the intercalated discs (IDs) that comprise the end-to-end connections between myocytes. IDs consist of three main junctional complexes: adherens junctions (AJs), gap junctions and desmosomes. Each of these junctional complexes is extremely important for maintenance of normal mechanical and electrical coupling between cardiomyocytes, and therefore for normal heart function. In the heart, the AJs are comprised predominantly of N-cadherin, which is highly expressed in the developing and mature myocardium. Classical cadherins are transmembrane proteins that mediate specific cell–cell adhesion. At the cytoplasmic side of the junction, either b-catenin or plakoglobin can interact directly with a core region within the C-terminus of the cadherin cytoplasmic domain. The N-terminal ortion of b-catenin or plakoglobin interacts with a-catenin, which links this complex to the cytoskeleton.
Wide known that intercellular adhesion is important for the development of any multicellular organism, but here we would like discuss other function of adherens junctions proteins in embryonic and adult heart. In our work we have focused on signalling function of b- and a-E – catenin. Beta –catenin is main transcriptional co-activator of canonical Wnt –signalling and in such way it involved in cells differentiation and proliferation control. Signalling function of a-E catenin recovered during last decades, it was shown that last one can regulate a few important signalling passway including canonical Wnt, Hippo, Hedgehog and NF-κB. Thus signalling function of b-catenin in heart is controversial and signalling function of a-E catenin is fur from understood.
With conditional knockout mice using we have focused on signalling function of a E-catenin an b-catenin function in heart development and maturation. We observed the adult heart maturation delay in mice with b-catenin haploinsufficiency. Moreover, we registered fetal genes up-regulation in mutant mice together with canonical Wnt down-regulation. Our data suggest that canonical Wnt and b-catenin is highly important for new born heart maturation and terminal differentiation of cardiomyocytes. We also recovered that even basal level of canonical Wnt activity is critically important for adult heart adaptation for stress. This data supported by other experiments where we registered heart enlarged hypertrophy and heart failure in mice with a-E catenin missing. In these mice we registered higher level of b-catenin and Yap dependent transcription.
Deep Learning in Medicine and Computational Biology
Deep learning has been shown to be extremely successful in various domains from image and voice recognition to beating humans at playing games and sorting waste. A major part of this advancement is due to rich datasets available to researches in those domains. In biology the amount and the complexity of data has increased dramatically, over the past decade, making biology and medicine suitable domains for applying deep learning. At the beginning, applying even simple networks on these large amounts of biological data provided a sophisticated advantage over canonical machine learning methods. Now, new ideas were adapted from rapidly developing artificial intelligence field in order to improve the performance of deep learning models across various biological tasks, such as: genomics, medical diagnostics and biological image analysis. In this talk, we will review some of these major recent advances and discuss their potential impact on the field. The talk is based on a review paper – Computational biology – deep learning by William Jones, Kaur Alasoo, Dmytro Fishman et al. (accepted).
tba
Are we ready for personalised medicine? – urban microbiome perspective
Answering biological and medical questions with genomic big data.
There is a group of biological and medical questions that can be answered or solved better with the use of big data approaches. The talk will present case studies of such ones: a prototype of genomic data warehouse and basics of big-data-driven nucleotide-precision RNA-seq analysis. They may lead to new types of insights, useful for medical research on the molecular level as well as in the clinical use. Such next-generation sequencing based genomics approaches are intended to bridge the gap between the large datasets and their useful interpretations in personalised medicine.
Species differences in receptor-operated TRPC4 channels
The ever-growing number of sequenced animal genomes provides a wealth of information concerning not only evolutionary relations, but also molecular structure-function relations. In this talk, our recent work on receptor-operated TRPC4 channel expressed in intestinal myocytes of rodents will be discussed in the context of the role of certain naturally occurring structural differences of these channel proteins, as well as the potential issues concerning using animal models for translational research.
Tailored bioinformatics tools for neuromuscular function & disease.
Most bioinformatics tools and data resources are designed to have wide application, whereas more tailored approaches can provide analyses that are adapted to specific pathologies or tissue types. We have developed several systems biology tools for neuromuscular research, accessible through the Sys-Myo website (http://sys-myo.rhcloud.com/).
These tools include: (i) CellWhere – when given a list of genes/proteins, CellWhere returns a network map of gene functional associations overlaid onto a schema of the cell and its subcellular compartments. CellWhere can be used to help formulate mechanistic hypotheses and to screen data for proteins that may be at specific cell compartments. (ii) Muscle Gene Sets – public transcriptomic data are used to create lists of genes that were differentially expressed across muscle experiments and pathologies, to be used for gene enrichment testing. (iii) The Dystrophin Interactome – this tool combines public functional association data with new proteomic experimental data on the binding partners of Dystrophin, enabling exploration of the interactions of this important muscle protein with other cellular components. (iv) MyoMiner – this database summarizes gene co-expression across previous microarray studies of muscle tissues and pathologies, potentially yielding a new empirically-derived definition of functional gene clusters in muscle.