This work focuses on the need for modeling and predicting adverse outcomes in immunotoxicology to improve nonclinical assessments of the safety of immunomodulatory therapies. The integrated approach includes, first, the adverse outcome pathway concept established in the toxicology field, and, second, the systems medicine disease map approach for describing molecular mechanisms involved in a particular pathology. The proposed systems immunotoxicology workflow is illustrated with chimeric antigen receptor (CAR) T cell treatment as a use case. To this end, the linear adverse outcome pathway (AOP) is expanded into a molecular interaction model in standard systems biology formats. Then it is shown how knowledge related to immunotoxic events can be integrated, encoded, managed, and explored to benefit the research community. The map is accessible online at https://imsavar.elixir-luxembourg.org via the MINERVA Platform for browsing, commenting, and data visualization. Our work transforms a graphical illustration of an AOP into a digitally structured and standardized form, featuring precise and controlled vocabulary and supporting reproducible computational analyses. Because of annotations to source literature and databases, the map can be further expanded to match the evolving knowledge and research questions.

Tools for predicting COVID-19 outcomes enable personalized healthcare, potentially easing the disease burden. This collaborative study by 15 institutions across Europe aimed to develop a machine learning model for predicting the risk of in-hospital mortality post-SARS-CoV-2 infection. Blood samples and clinical data from 1286 COVID-19 patients collected from 2020 to 2023 across four cohorts in Europe and Canada were analyzed, with 2906 long non-coding RNAs profiled using targeted sequencing. From a discovery cohort combining three European cohorts and 804 patients, age and the long non-coding RNA LEF1-AS1 were identified as predictive features, yielding an AUC of 0.83 (95% CI 0.82–0.84) and a balanced accuracy of 0.78 (95% CI 0.77–0.79) with a feedforward neural network classifier. Validation in an independent Canadian cohort of 482 patients showed consistent performance. Cox regression analysis indicated that higher levels of LEF1-AS1 correlated with reduced mortality risk (age-adjusted hazard ratio 0.54, 95% CI 0.40–0.74). Quantitative PCR validated LEF1-AS1’s adaptability to be measured in hospital settings. Here, we demonstrate a promising predictive model for enhancing COVID-19 patient management.

Information Technology in agriculture is intended to provide knowledge to the farmers in order to make more informed and valuable decisions. In this paper we present an Integrated Agriculture Information Framework (IAIF) to enable knowledge extraction from multiple domain related repositories. IAIF has been designed and developed with the core aim of meaningfully representing combining, merging and aggregating the data present in existing knowledge repositories through the use of metadata and domain ontologies. The domain ontology serves to link the resources with the domain knowledge with the help of respective metadata. A key feature of the semantics based approach adopted is that ontology also serves as the basis for capturing the needed information for accessing different types of data sources. We also present an Ontology engineering methodology for the domain that allows for formalized modeling of domain schema and resource knowledge. In addition, we discuss the methodology for linking multiple types of knowledge resources.

The data aggregation process can be considered a problem of multi-objective optimization which reduces the size of data in such a way that its relevance with original data remains as closer as possible. Data Aggregation is of great importance in Wireless Sensor Networks, Vehicular Ad-hoc Networks to transmit the recorded data in time over low bandwidth. In this regard, data aggregation solutions have been developed; however, their actual usage has been limited, for the reason of low accuracy and high processing time. In this paper, particle swarm optimization (PSO) is used to optimize process of multi-objective data aggregation in vehicular ad-hoc network. In our work processing time for aggregation and aggregation quality have been set as objectives. The proposed method has been compared with state of the art existing aggregation techniques. Experimental results show that our method simplifies aggregation effectively and obtains a higher aggregation accuracy compared to the other data aggregation methods.

Recently, content-based Spam detection frameworks are receiving a significant amount of attention by academic researchers and industrial practitioners. However, the anticipated wide scale proliferation is limited (mainly) because of two important shortcomings: (1) high false alarm rate that results in moving legitimate messages into Spam folders, and (2) inability to self-learn a user's profile, as a result, they are unable to identify useful Spam (we call it Good Spam) that might be of great interest to a user's personal or business aspirations. In this paper, we propose USpam, a system that uses ontologies to model features that are extracted rom a user's profile. The features are given to machine learning classifiers J48 and Naive Bayes -- that learn a user centric model of Good Spam or Bad Spam. As a result, the system puts a message into a user's inbox if its contents are relevant to his interests. The USpam is evaluated on NRON Spam datasets, and the results of experiments reveal that false alarms are reduced by 10% to 30% compared with existing prior art without compromising the detection accuracy.