11259403 OR Neural network model of gene expression Dotaz Zobrazit nápovědu
BACKGROUND: One possible approach how to economically facilitate gene expression profiling is to use the L1000 platform which measures the expression of ∼1,000 landmark genes and uses a computational method to infer the expression of another ∼10,000 genes. One such method for the gene expression inference is a D-GEX which employs neural networks. RESULTS: We propose two novel D-GEX architectures that significantly improve the quality of the inference by increasing the capacity of a network without any increase in the number of trained parameters. The architectures partition the network into individual towers. Our best proposed architecture - a checkerboard architecture with a skip connection and five towers - together with minor changes in the training protocol improves the average mean absolute error of the inference from 0.134 to 0.128. CONCLUSIONS: Our proposed approach increases the gene expression inference accuracy without increasing the number of weights of the model and thus without increasing the memory footprint of the model that is limiting its usage.
Gene expression profiling was made more cost-effective by the NIH LINCS program that profiles only ∼1, 000 selected landmark genes and uses them to reconstruct the whole profile. The D-GEX method employs neural networks to infer the entire profile. However, the original D-GEX can be significantly improved. We propose a novel transformative adaptive activation function that improves the gene expression inference even further and which generalizes several existing adaptive activation functions. Our improved neural network achieves an average mean absolute error of 0.1340, which is a significant improvement over our reimplementation of the original D-GEX, which achieves an average mean absolute error of 0.1637. The proposed transformative adaptive function enables a significantly more accurate reconstruction of the full gene expression profiles with only a small increase in the complexity of the model and its training procedure compared to other methods.
Vlastnosti neuronálních okruhů závisejí na počtu funkčních elementů (neuronů), četnosti interneuronálních spojení a funkční charakteristice synapsí. Komplexita neuronálních okruhů podmiňuje kvalitativní rozdíly v integrační činnosti mozku (úroveň inteligence a kognitivních funkcí, kapacita paměti). Vývoj nervového systému je jen částečně determinován geneticky (nature), významněji je ovlivňován vlivy prostředí (nurture). I faktory prostředí mohou modulovat vývoj struktury a její funkci zásahem do transkripce genů. Takto je například řízena exprese specifických signálních látek buněčných povrchů, tzv. adhezivních molekul, které poskytují informace pro růstové vrcholy axonů a dendritů. Podobně je řízena exprese neurotrofních faktorů, které napomáhají udržovat vzájemné funkční vztahy, případně stimulovat jejich další diferenciaci. Kromě přítomnosti či nepřítomnosti určitých vloh (genů), potřebných pro optimalizaci struktury a funkce neuronálních okruhů, je nutné uvažovat i o faktorech, které umožní, aby se tyto vlohy uplatnily. Vedle biologických faktorů, jako je úroveň nutrice, zásobení kyslíkem, případně úrazy a onemocnění, je pro vyladění funkce neuronálních okruhů nutná i adekvátní stimulace a zpracovávání přijatých informací. Hlavní zdroj stimulačních signálů představuje pro člověka jazyk, sociální prostředí a kultura.
Features of neuronal circuits depend on the number of functional elements (neurons), density of interneuronal connections and functional characteristics of synapses. Complexity of neuronal circuits determines the qualitative differences in the integration activity of the brain (level of intelligence and cognitive functions, memory capacity). Nerve system development is only partially determined by genetic factors (nature); probably more significant is the effect of environment (nurture). Environmental factors can interfere with gene transcription and thus modulate the development of structure and function. It is the mechanism controlling the expression of specific signals at neuronal surfaces, which include adhesive molecules providing information for growing cones of axons and dendrites. The expression of neurotrophic factors which help to maintain mutual functional relations between neurons is controlled similarly. Beside the presence or absence of certain dispositions (genes), necessary for the optimal development of the structure and function of neuronal circuits, also the factors which carry out the inherited dispositions have to be considered. Beside the biological factors like nutrition, oxygen supply or injuries and diseases, the adequate stimulation and information processing are necessary for the functional tuning of neuronal circuits. The principal source of stimulatory signals for humans represents the language, social environment and culture.
Ionotropní glutamátové receptory NMDA typu hrají podstatnou roli v procesu synaptické plasticity, která je základem učení a paměti. Změna spojení mezi neurony je důsledkem působení jednak lokálních faktorů v synapsi a jednak celkových faktorů, které zahrnují aktivaci proteinkináz a přenos signálu do buněčného jádra, kde proteinkinázy fosforylují konstitutivně exprimované transkripční faktory, které řídí expresi tzv. genů časné odpovědi, z nichž některé jsou samy transkripční faktory a kontrolují expresi dalších cílových genů. Článek prezentuje dosud známé poznatky o mechanizmu přenosu signálu od NMDA receptoru k indukci modelového genu časné odpovědi c-fos, jehož exprese je pod kontrolou několika konstitutivních faktorů, mezi jinými CREB (cAMP-response element binding protein), který musí být fosforylován na kritickém Ser133, aby aktivoval transkripci. Rychlým důsledkem aktivace NMDA receptoru je translokace komplexu Ca2+-kalmodulin do jádra a aktivace Ca2+/kalmodulin dependentní kinázy IV. S pomalejší kinetikou dochází k aktivaci Ras-MAPK (mitogen-activated protein kinase) kinázové kaskády, která je zřejmě v signalizaci do jádra neuronu nejdůležitější a přes jiné specifické signální molekuly integruje i signalizaci využívající Ca2+ nebo cyklický AMP jako druhého posla.
Ionotropic NMDA glutamate receptors play a central role in the process of synaptic plasticity underlying learning and memory. Alteration of neuronal connections results from local effects in the target synapse as well as from the whole-cell factors that encompass activation of protein kinases and signal transduction to the cell nucleus, where the kinases phosphorylate constitutively expressed transcription factors controlling expression of a set of immediate-early genes; some of them, in turn, are themselves transcription factors and regulate expression of further target genes. The article presents some of knowledge accumulated to date, on the mechanism of signal transduction from NMDA receptor to induction of the model immediate-early gene c-fos, whose expression is controlled by several constitutive factors, among them by the CREB (cAMP-response element binding protein), that must be phosphorylated at a critical Ser133, in order to activate transcription. One fast event in response to NMDA receptor activation is translocation of the Ca2+-calmodulin complex to the cell nucleus and activation of Ca2+/calmodulin-dependent protein kinase IV. With slower kinetics, the Ras-MAPK (mitogen-activated protein kinase) kinase cascade is activated, emerging in the signalling to the neuronal nucleus as the most significant pathway, through other specific signalling molecules integrating also signals originally using Ca2+ or cyclic AMP as the second messengers.
The involvement of microRNAs (miRNAs) in orchestrating self-renewal and differentiation of stem cells has been revealed in a number of recent studies. And while in human pluripotent stem cells, miRNAs have been directly linked to the core pluripotency network, including the cell cycle regulation and the maintenance of the self-renewing capacity, their role in the onset of differentiation in other contexts, such as determination of neural cell fate, remains poorly described. To bridge this gap, we used three model cell types to study miRNA expression patterns: human embryonic stem cells (hESCs), hESCs-derived self-renewing neural stem cells (NSCs), and differentiating NSCs. The comprehensive miRNA profiling presented here reveals novel sets of miRNAs differentially expressed during human neural cell fate determination in vitro. Furthermore, we report a miRNA expression profile of self-renewing human NSCs, which has been lacking to this date. Our data also indicates that miRNA clusters enriched in NSCs share the target-determining seed sequence with cell cycle regulatory miRNAs expressed in pluripotent hESCs. Lastly, our mechanistic experiments confirmed that cluster miR-17-92, one of the NSCs-enriched clusters, is directly transcriptionally regulated by transcription factor c-MYC.
Motivation: Whole genome expression profiling of large cohorts of different types of cancer led to the identification of distinct molecular subcategories (subtypes) that may partially explain the observed inter-tumoral heterogeneity. This is also the case of colorectal cancer (CRC) where several such categorizations have been proposed. Despite recent developments, the problem of subtype definition and recognition remains open, one of the causes being the intrinsic heterogeneity of each tumor, which is difficult to estimate from gene expression profiles. However, one of the observations of these studies indicates that there may be links between the dominant tumor morphology characteristics and the molecular subtypes. Benefiting from a large collection of CRC samples, comprising both gene expression and histopathology images, we investigated the possibility of building image-based classifiers able to predict the molecular subtypes. We employed deep convolutional neural networks for extracting local descriptors which were then used for constructing a dictionary-based representation of each tumor sample. A set of support vector machine classifiers were trained to solve different binary decision problems, their combined outputs being used to predict one of the five molecular subtypes. Results: A hierarchical decomposition of the multi-class problem was obtained with an overall accuracy of 0.84 (95%CI=0.79-0.88). The predictions from the image-based classifier showed significant prognostic value similar to their molecular counterparts. Contact: popovici@iba.muni.cz. Availability and Implementation: Source code used for the image analysis is freely available from https://github.com/higex/qpath . Supplementary information: Supplementary data are available at Bioinformatics online.
- MeSH
- kolorektální nádory diagnóza genetika metabolismus patologie MeSH
- lidé MeSH
- nádorové biomarkery * MeSH
- neuronové sítě (počítačové) * MeSH
- počítačové zpracování obrazu metody MeSH
- prognóza MeSH
- regulace genové exprese u nádorů MeSH
- support vector machine MeSH
- Check Tag
- lidé MeSH
- Publikační typ
- časopisecké články MeSH
Major advances in the genomics and epigenomics of diffuse gliomas and glioblastoma to date have not been translated into effective therapy, necessitating pursuit of alternative treatment approaches for these therapeutically challenging tumors. Current knowledge of microtubules in cancer and the development of new microtubule-based treatment strategies for high-grade gliomas are the topic in this review article. Discussed are cellular, molecular, and pharmacologic aspects of the microtubule cytoskeleton underlying mitosis and interactions with other cellular partners involved in cell cycle progression, directional cell migration, and tumor invasion. Special focus is placed on (1) the aberrant overexpression of βIII-tubulin, a survival factor associated with hypoxic tumor microenvironment and dynamic instability of microtubules; (2) the ectopic overexpression of γ-tubulin, which in addition to its conventional role as a microtubule-nucleating protein has recently emerged as a transcription factor interacting with oncogenes and kinases; (3) the microtubule-severing ATPase spastin and its emerging role in cell motility of glioblastoma cells; and (4) the modulating role of posttranslational modifications of tubulin in the context of interaction of microtubules with motor proteins. Specific antineoplastic strategies discussed include downregulation of targeted molecules aimed at achieving a sensitization effect on currently used mainstay therapies. The potential role of new classes of tubulin-binding agents and ATPase inhibitors is also examined. Understanding the cellular and molecular mechanisms underpinning the distinct behaviors of microtubules in glioma tumorigenesis and drug resistance is key to the discovery of novel molecular targets that will fundamentally change the prognostic outlook of patients with diffuse high-grade gliomas.
- MeSH
- antitumorózní látky farmakologie terapeutické užití MeSH
- gliom farmakoterapie genetika metabolismus MeSH
- karcinogeneze účinky léků genetika metabolismus MeSH
- lidé MeSH
- mikrotubuly účinky léků genetika metabolismus MeSH
- neuronové sítě (počítačové) MeSH
- regulace genové exprese u nádorů účinky léků genetika MeSH
- zvířata MeSH
- Check Tag
- lidé MeSH
- zvířata MeSH
- Publikační typ
- časopisecké články MeSH
- přehledy MeSH
Autism spectrum disorder (ASD) is a neurodevelopmental disorder accompanied by narrow interests, difficulties in communication and social interaction, and repetitive behavior. In addition, ASD is frequently associated with eating and feeding problems. Although the symptoms of ASD are more likely to be observed in boys, the prevalence of eating disorders is more common in females. The ingestive behavior is regulated by the integrative system of the brain, which involves both homeostatic and hedonic neural circuits. Sex differences in the physiology of food intake depend on sex hormones regulating the expression of the ASD-associated Shank genes. Shank3 mutation leads to ASD-like traits and Shank3B -/- mice have been established as an animal model to study the neurobiology of ASD. Therefore, the long-lasting neuronal activity in the central neural circuit related to the homeostatic and hedonic regulation of food intake was evaluated in both sexes of Shank3B mice, followed by the evaluation of the food intake and preference. In the Shank3B +/+ genotype, well-preserved relationships in the tonic activity within the homeostatic neural network together with the relationships between ingestion and hedonic preference were observed in males but were reduced in females. These interrelations were partially or completely lost in the mice with the Shank3B -/- genotype. A decreased hedonic preference for the sweet taste but increased total food intake was found in the Shank3B -/- mice. In the Shank3B -/- group, there were altered sex differences related to the amount of tonic cell activity in the hedonic and homeostatic neural networks, together with altered sex differences in sweet and sweet-fat solution intake. Furthermore, the Shank3B -/- females exhibited an increased intake and preference for cheese compared to the Shank3B +/+ ones. The obtained data indicate altered functional crosstalk between the central homeostatic and hedonic neural circuits involved in the regulation of food intake in ASD.
- MeSH
- homeostáza * fyziologie MeSH
- mikrofilamentové proteiny * genetika metabolismus MeSH
- modely nemocí na zvířatech MeSH
- myši inbrední C57BL MeSH
- myši knockoutované MeSH
- myši MeSH
- pohlavní dimorfismus * MeSH
- poruchy autistického spektra * genetika metabolismus MeSH
- preference v jídle fyziologie MeSH
- přijímání potravy * fyziologie genetika MeSH
- proteiny nervové tkáně * genetika MeSH
- protoonkogenní proteiny c-fos metabolismus biosyntéza MeSH
- zvířata MeSH
- Check Tag
- mužské pohlaví MeSH
- myši MeSH
- ženské pohlaví MeSH
- zvířata MeSH
- Publikační typ
- časopisecké články MeSH
