Institut für Informatik und Computational Science
Refine
Document Type
- Conference Proceeding (4)
- Doctoral Thesis (2)
- Article (1)
Keywords
- Informatikdidaktik (2)
- Antwortmengenprogrammierung (1)
- Assessment (1)
- Bildung (1)
- Computer Science Education (1)
- Constraint Solving (1)
- Deduction (1)
- Didaktische Konzepte (1)
- Diversität (1)
- HDI (1)
- Heterogenität (1)
- Hochschuldidaktik (1)
- Hochschullehre (1)
- ICT (1)
- Informatics (1)
- Informatik (1)
- Informatikstudium (1)
- Key Competencies (1)
- Lebenslanges Lernen (1)
- Logic Programming (1)
- Logics (1)
- Optimization (1)
- Planing (1)
- Schlüsselkompetenzen (1)
- Segmentierung (1)
- Systembiologie (1)
- answer set programming (1)
- anti-cancer drugs (1)
- bild (1)
- deep neural networks (1)
- digitale Hochschullehre (1)
- drug-sensitivity prediction (1)
- education (1)
- image (1)
- logical signaling networks (1)
- logische Signalnetzwerke (1)
- medical (1)
- medizinisch (1)
- segmentation (1)
- systems biology (1)
Institute
- Extern (7) (remove)
Large-scale databases that report the inhibitory capacities of many combinations of candidate drug compounds and cultivated cancer cell lines have driven the development of preclinical drug-sensitivity models based on machine learning. However, cultivated cell lines have devolved from human cancer cells over years or even decades under selective pressure in culture conditions. Moreover, models that have been trained on in vitro data cannot account for interactions with other types of cells. Drug-response data that are based on patient-derived cell cultures, xenografts, and organoids, on the other hand, are not available in the quantities that are needed to train high-capacity machine-learning models. We found that pre-training deep neural network models of drug sensitivity on in vitro drug-sensitivity databases before fine-tuning the model parameters on patient-derived data improves the models’ accuracy and improves the biological plausibility of the features, compared to training only on patient-derived data. From our experiments, we can conclude that pre-trained models outperform models that have been trained on the target domains in the vast majority of cases.
Preface
(2010)
The workshops on (constraint) logic programming (WLP) are the annual meeting of the Society of Logic Programming (GLP e.V.) and bring together researchers interested in logic programming, constraint programming, and related areas like databases, artificial intelligence and operations research. In this decade, previous workshops took place in Dresden (2008), Würzburg (2007), Vienna (2006), Ulm (2005), Potsdam (2004), Dresden (2002), Kiel (2001), and Würzburg (2000). Contributions to workshops deal with all theoretical, experimental, and application aspects of constraint programming (CP) and logic programming (LP), including foundations of constraint/ logic programming. Some of the special topics are constraint solving and optimization, extensions of functional logic programming, deductive databases, data mining, nonmonotonic reasoning, , interaction of CP/LP with other formalisms like agents, XML, JAVA, program analysis, program transformation, program verification, meta programming, parallelism and concurrency, answer set programming, implementation and software techniques (e.g., types, modularity, design patterns), applications (e.g., in production, environment, education, internet), constraint/logic programming for semantic web systems and applications, reasoning on the semantic web, data modelling for the web, semistructured data, and web query languages.
Deciphering the functioning of biological networks is one of the central tasks in systems biology. In particular, signal transduction networks are crucial for the understanding of the cellular response to external and internal perturbations. Importantly, in order to cope with the complexity of these networks, mathematical and computational modeling is required. We propose a computational modeling framework in order to achieve more robust discoveries in the context of logical signaling networks. More precisely, we focus on modeling the response of logical signaling networks by means of automated reasoning using Answer Set Programming (ASP). ASP provides a declarative language for modeling various knowledge representation and reasoning problems. Moreover, available ASP solvers provide several reasoning modes for assessing the multitude of answer sets. Therefore, leveraging its rich modeling language and its highly efficient solving capacities, we use ASP to address three challenging problems in the context of logical signaling networks: learning of (Boolean) logical networks, experimental design, and identification of intervention strategies. Overall, the contribution of this thesis is three-fold. Firstly, we introduce a mathematical framework for characterizing and reasoning on the response of logical signaling networks. Secondly, we contribute to a growing list of successful applications of ASP in systems biology. Thirdly, we present a software providing a complete pipeline for automated reasoning on the response of logical signaling networks.
We introduce a simple approach extending the input language of Answer Set Programming (ASP) systems by multi-valued propositions. Our approach is implemented as a (prototypical) preprocessor translating logic programs with multi-valued propositions into logic programs with Boolean propositions only. Our translation is modular and heavily benefits from the expressive input language of ASP. The resulting approach, along with its implementation, allows for solving interesting constraint satisfaction problems in ASP, showing a good performance.