Project publications

Abstract. Agricultural practices based on sustainable principles (e.g., conservation tillage, crop rotation, cover cropping, and the application of organic inputs) have been tested for their potential to improve soil structure, enhance soil organic matter, and support agrobiodiversity. These practices are directly linked to soil microbial diversity. Diverse soil microbial communities play multiple roles in promoting beneficial interactions between plants and their environment and in maintaining functional agroecosystems. Key functions enabled by soil microorganisms are carbon dynamics, nutrient cycling, soil structure improvement, pathogen suppression, plant growth promotion, and stress tolerance. Soil microorganisms are increasingly recognized as a promising but still underexploited source in tackling sustainability challenges in agricultural production. However, their potential varies depending on the interactions among abiotic and biotic factors, as well as the applied cultivation practices. In recent decades, advances in DNA extraction from soil and next-generation sequencing (NGS) technologies have enabled comprehensive characterization of microbial diversity, community composition, and functional potential for assessing soil health and agroecosystem functioning. Understanding, predicting, and exploring relevant plant–soil–microbiome interactions are essential for enhancing agroecosystem capacity for sustainable production. This review paper highlights how different factors and agricultural practices affect microbiome biodiversity. The focus is on microbiome approaches that integrate information on community composition with assessments of functional potential and measured microbial activity and ecosystem processes, combining state-of-the-art molecular monitoring, ecological indicators, and predictive modeling to support evidence-based management recommendations, distinguishing approaches that are currently applicable in agricultural practice from those that require further experimental validation.

Abstract. Biostimulants based on Bradyrhizobium japonicum are commonly used in soybean production. However, the effect of nitrogen-fixing bacteria in consortia with other plant growth-promoting rhizobacteria (PGPR) and their integration with mineral nutrients on soybean seed quality has not been explored. The study aimed to examine the effects of five treatments on seed germination and initial seedling growth of two soybean cultivars (‘NS Apolo’, ‘NS Rubin’): control (untreated seeds); Br. japonicum (BJ), BJ and nutrient complex (NC), BJ, Azotobacter chroococcum (AC), Bacillus subtilis (BS), and NC; BJ, AC, Bacillus megaterium (BM), and NC. Seed treatments significantly enhanced germination energy, seedling vigor index, root length, fresh shoot weight, fresh root weight, dry shoot weight, and dry root weight of both cultivars, as well as final germination, shoot length, and shoot elongation rate of ‘NS Rubin’, as compared to the control. The highest effect on the investigated parameters was achieved by integrated use of PGPR and nutrients (BJ + BM + AC + NC), indicating that integration of PGPR with a targeted NC represents an innovative approach with practical implications for improving early soybean establishment and field performance.

Abstract. The EU-funded ECO-READY project will ensure that European farmers and society interests are reflected in future policymaking and monitoring with its real-time surveillance system. ECOREADY develops a central digital infrastructure (Observatory) that aggregates all existing data, models and knowledge on food security in the context of impact of climate change on agrobiodiversity and invasive species and genetic resources, sustainable productivity, animal and plant diseases and nutritional composition. The platform will include policy recommendations, contingency plans and resilience strategies developed by 10 Living Labs and provide real-time assessments for the food system, with forecasts updated frequently and consistently. ECOREADY will ensure the European society ‘s interests to be reflected in future policy making and monitoring, having them actively engaged to shaping policy recommendations from an early stage. That will be facilitated by the knowledge communicated via the digital...

Abstract. With its real-time surveillance system, the EU-funded ECO-READY project will enable that the challenges and interests of farmers and society at theEuropean level are represented in future policymaking and monitoring. ECO-READY develops an Observatory (digital infrastructure) that brings together all existing data, models and knowledge on food security in the context of impact of the climate change on agro-biodiversity and invasive species and genetic resources, sustainable production, animal and plant diseases, and nutritional composition. With forecasts updated regularly and continuously, the observatory will offer real-time assessments for the food system along with policy recommendations, contingency plans, and resilience solutions created by 10 Living Labs. The data will be communicated via the digital observatory, informing citizens and supporting policy makers and agri-food supply chain stakeholders to make evidence-based decisions. In addition to the development of a digital observatory, ECO-READY is dedicated to scenario development (short- and long- term) activities that explore agricultural practices, policy changes, innovative farming ideas, and consumer-producer interactions to help mitigate future shocks. In order to determine their regional food priorities, the ten ECO-READY living labs covering the majority of Europe's geo-climatic zones have chosen important drivers. Taking into account drivers, for each of the living labs tailored-made scenarios relevant to their region and selected priority crops were created. ECO-READY, in its intention of solving the climate-food-biodiversity challenges, involved all relevant stakeholders for working together across different fields and taking real action that includes everyone and leads to lasting change

Abstract. The legume–rhizobia symbiotic N2 fixation is considered the most efficient and important process in crop production. Besides their potential to reduce atmospheric nitrogen, rhizobia accomplish various PGP (Plant Growth Promoting) activities such as phosphate solubilization, siderophore and phytohormone production, thus improving nutrient availability, soil structure and fertility, and crop productivity. Utilizing rhizobial inoculants is a low-cost and highly efficient approach that reduces inputs of synthetic fertilizers while promoting sustainable agricultural and environmental stability. To improve inoculation success, it is crucial to continuously investigate native rhizobia populations as an inexhaustible source of superior strains with an enhanced adaptability to specific agricultural environments. This research aims to characterize Bradyrhizobium strains with high plant growth-promoting properties for potential use as bioinoculants for soybean. Bradyrhizobium japonicum strains (33) were obtained from the collection of the Laboratory for Microbiology of the Institute of Field and Vegetable Crops. These strains were originally isolated from root nodules collected from different soybean cultivars and locations in Vojvodina province. The qualitative estimation of Bradyrhizobium strains for phosphate solubilization and mineralization ability and siderophore production was performed in vitro on appropriate media. The Salkowski reagent colorimetric method was used for quantitative determination of indole-3-acetic acid (IAA). Based on PGP traits, 10 strains were selected for the inoculation experiment in pots to examine the effect on soybean root nodulation in soils of high and low fertility. In a full bloom stage of soybean, the number and dry weight of nodules were assessed. This study revealed that all strains synthesize IAA, with concentrations ranging from 8.97 μg/mL to 41.17 μg/mL. Isolates Bj2, Bj4, Bj6, and Bj8 produced the highest amount of IAA. Among tested strains, 12 showed phosphate mineralization and only five phosphate solubilization abilities. However, the solubilization and mineralization activities varied, with the highest diameter of the halo zones formed in strains Bj1, Bj4, Bj7, Bj18, and Bj24. The appearance of yellow-orange zones indicated siderophore production in 16 Bradyrhizobium strains, and the best producers were Bj1, Bj2, Bj6, Bj12, and Bj21. All strains were able to successfully induce nodulation on soybean roots in both high- and low-fertility soils. However, inoculation with Bj1, Bj2, Bj4, Bj6, Bj7, and Bj8 achieved the highest score according to the nodulation scale. Although this research confirmed promising PGP and symbiotic potentials of indigenous Bradyrhizobium strains, extensive field studies are required to evaluate their efficiency for further bioinoculant development.

Abstract. The objective of this study was to analyze the response of antioxidant parameters in soybean plants inoculated with newly isolated Bradyrhizobium japonicum and Bacillus subtilis strains as single and co-inoculants under drought stress. Bacterial strains were selected according to osmotic stress tolerance (in the presence of 36% PEG 6000) in appropriate liquid media. The effect of soybean inoculation was examined in a soil pot experiment in water deficit conditions (0 and 7 days withholding water). The influence of water stress and inoculation was evaluated in soybean leaves, roots, and nodules through guaiacol peroxidase (POX), ionically cell-wall-bound peroxidase (POD) activity, and ABTS˙+ radical cation scavenging capacity, as well as parameters of N-fixation efficiency. The results showed a significant influence of inoculation on constitutive and drought-induced antioxidant and N-fixation parameters. Inoculation increased the activity of POX (up to 116, 169, and 245%), POD (up to 116, 102, and 159%), and antioxidant capacity (up to 74, 76, and 81%) in soybean leaves, roots, and nodules under water deficit, respectively. Application of bacterial strains resulted in higher shoot, root, and nodule weight and nitrogen content both in non-stressed and drought stress conditions. Overall, co-inoculation had better effects on the investigated soybean parameters compared to single inoculation. Selection and application of bacterial strains with improved tolerance to drought stress is necessary in developing inoculants that would result in enhanced crop production under unfavorable environmental conditions.

Abstract. Climate-resilient food systems require simultaneously addressing climate mitigation, food security, and biodiversity conservation. Yet most policy frameworks treat these interdependent goals as separate challenges, fragmenting action across sectors. This study examines how diverse food system actors (producers, policymakers, civil society, consumers) reach consensus on integrated resilience strategies and governance architectures. A two-round Delphi with SWOT/TOWS analysis integrated 79 stakeholders from four groups across 14 EU countries. Participants consistently favored a governance model that couples binding EU-level standards with flexibility in regional implementation. The resilience strategies receiving strongest support centered on production diversification, traceability along supply chains, interactor knowledge networks, and digital monitoring systems. Northern Europe emphasizes technological innovation, whereas Southern Europe emphasizes social capacity-building, revealing complementary implementation pathways. This pattern, where broad agreement on objectives coexists with divergence on implementation pathways, points to a key design challenge for climate governance: how to maintain policy ambition at the European level without ignoring the real differences in capacity and institutional context across member states. The study demonstrates how stakeholder-consensus governance can simultaneously achieve climate resilience, address power asymmetries, and ensure institutional feasibility in food system transformation.

Abstract. With the projected increase in the global population and food demand on a planet in climate crisis, the debate about the role of livestock continues to intensify. Animals farmed for food are both a victim and a contributor of climate change (CC), with animals suffering from heat stress due to CC but also a major contributor to greenhouse gas emissions (GHG). Livestock studies have demonstrated that novel breeding technologies could partly mitigate these negative effects, yet their wider implications on food security and the socio-economy are not well investigated yet. In this proof-of-concept study we incorporate, 1) the effects of heat stress on livestock’s productivity and feed efficiency, and 2) the use of genomic selection as a recently established novel breeding technology, into a global macroeconomic projection model (GMPM) to evaluate their broader impacts on food security. We find that both effects can drastically alter the model projections. In particular, the results suggest that novel breeding technologies can substantially mitigate adverse effects of CC on production, food availability, land use, and GHG, whilst also causing implications for the labour market. The study demonstrates the importance of integrating advances in livestock science and technology into GMPMs for better projections that inform livestock management, policy and consumer decision-making, requiring closer collaboration between livestock scientists and systems modellers.

Abstract. The amount of software engineering data is constantly growing, as more and more developers employ online services to store their code, keep track of bugs, or even discuss issues. The data residing in these services can be mined to address different research challenges; therefore, certain initiatives have been established to encourage sharing research datasets collecting them. In this work, we investigate the effect of such an initiative; we create a directory that includes the papers and the corresponding datasets of the data track of the Mining Software Engineering (MSR) conference.

Abstract. Agriculture is an essential factor in the climate-food-biodiversity nexus, affecting climate resilience and, ulti- mately, whether we achieve the global Sustainable Development Goals. This study aimed to investigate the local- global interaction by integrating sustainability indicators with three modelling methods to address agriculture and climate resilience. The novelty of this research lies in its innovative methodology, which employs locally sourced indicators to integrate with climate and crop growth models (DSSAT, local biophysics), fed into large- scale equilibrium models (MAGNET, global economics) and Life-Cycle Assessment tools (LCA, environmental feedback). This enables precise mapping and analyzing regions that are most and least vulnerable to climate change, which is crucial for informing policymakers. Additionally, the novel methodology has incorporated focus groups to design a set of indicators that are compatible with typological input data for the modelling protocol.
Our methodology quantified the impact of heat, drought, CO2, and extreme weather conditions on local yield changes. This approach uniquely combined regional-level data with five types of indicators: farming practices, water, climate/soil, biodiversity, and economics. Focus groups were instrumental in the process of gathering, selecting, and fine-tuning indicators, identifying gaps, and areas where policies should be tailored and targeted.
This innovative work represents a significant step forward for evidence-based policy-making and allows us to emphasize the role of “local-to-global” feedback in scaling up models. It demonstrates how localized climate extremes can disproportionately influence the stability of global wheat production. It is important to highlight that biodiversity indicators are significantly missing from the large-scale modelling of the climate-food-biodiversity nexus.

Abstract. Integrated modelling of climate change impacts in the soil–plant–atmosphere system can improve agroecosystem resilience and resource use efficiency. For the research on tomato (Solanum lycopersicum L.), bell sweet peppers (Capsicum annuum L.), and winter oilseed rape (Brassica napus L.), the CSM-CROPGRO-Tomato, CSM-CROPGRO-Papper and CSM-CROPGROCanola models were used as an example for Central European environment, including the Czech Republic. Their outputs were validated using an extensive experimental field dataset with contrasting climates and soils. The CSM-CROPGROCanola model has tested the three-winter oilseed rape (WOSR) varieties from three experimental sites. To identify the most promising agronomic adaptation strategies for thermophilic vegetables (VEG) in open-field growing conditions, the ability of the CSM-CROPGRO-Tomato and CROPGRO-Papper models have been tested. To assess the risk of compound climate/ weather events (CEs) for WOSR production and the most promised areas with technical ripening of fruit tomatoes and sweet peppers, the ensemble of the Regional Climate Models for 2001–2020, 2021–2040 and 2041–2060 have been used. We found that linking the CSM-CROPGRO system with an ensemble of climate models is a reliable tool for testing new varieties of WOSR and VEG and assessing their suitability for future climate change.

Abstract. European policies on climate change (CC), food security (FS), and biodiversity (BD) represent the EU’s commitment to a sustainable agri-food system, highlighting the interdependence between environmental health and food security. By analyzing key drivers and indicators, the present study evaluates the effectiveness of existing measures and identifies gaps in the policy framework. A Scoping Group activity facilitated dialogue between policymakers, industry, and farmer representatives to gather feedback and strengthen the data–policy link. The results highlight progress in areas such as promoting sustainable agriculture and biodiversity, while pointing out unresolved issues like the challenges faced by smallholder farmers. The study emphasizes the need for real-time monitoring tools and tailored solutions to address the complexities of the agri-food system. It also encourages the integration of emerging technologies, such as IoT and AI, to enhance the sustainability of agricultural practices. Ultimately, the findings call for a landscape-specific approach to maximize biodiversity gains, mitigate climate impacts, and ensure food security within the broader context of the EU’s ecological and socio-economic challenges.

Abstract. Artificial intelligence may not just provide information services but could also have a major impact on the organization of planning processes, adding higher levels of flexibility to efficiency. Apart from AI’s information services, the added value of flexibility may especially support processes that are confronted with high variabilities as, e.g. processes dealing with risk management and contingency planning.

Abstract. The livestock system in Europe relies on a complex holistic equilibrium that is the outcome of an interplay of demand, market, crop production, livestock production, land use, water availability, and other factors. When modeling future scenarios of water consumption by livestock systems, the most suitable tools result from the interconnectivity of growth models, economic models, and climate models. We integrated the Environmental Policy Integrated Climate growth model (EPIC), animal-level model (RUMINANT), economic model (Global Biosphere Management Model, GLOBIOM), EURO-CORDEX climate models, and regression models. This study developed novel livestock production scenarios for individual regions of the Czech Republic with estimations of the categories of livestock that have been bred during the last 20 years and will be bred in the future and what their water consumption will be, both throughout the year and in particular seasons. First, the numbers of farm animals, namely, cattle, pigs, sheep, horses, goats, and poultry in 2002–2020 were evaluated, and their numbers were predicted for the following years until 2050. Second, livestock water consumption per region was determined based on the number of livestock individuals. Third, changes in the amount of water consumed by livestock per year in individual regions in 2050 compared to 2005 were estimated.

Abstract. The availability of code snippets in online repositories like GitHub has led to an uptick in code reuse, this way further supporting an open-source component-based development paradigm. The likelihood of code reuse rises when the code components or snippets are of high quality, especially in terms of readability, making their integration and upkeep simpler. Toward this direction, we have developed a dataset of code snippets that takes into account both the functional and the quality characteristics of the snippets. The dataset is based on the CodeSearchNet corpus and comprises additional information, including static analysis metrics, code violations, readability assessments, and source code similarity metrics. Thus, using this dataset, both software researchers and practitioners can conveniently find and employ code snippets that satisfy diverse functional needs while also demonstrating excellent readability and maintainability.

Abstract. Nowadays developers search online for reusable solutions to their problems in the form of source code snippets. As this paradigm can greatly reduce the time and effort required for software development, several systems have been proposed to automate the process of finding reusable snippets. However, contemporary systems also have certain limitations; several of them do not support queries in natural language and/or they only output API calls, thus limiting their ease of use. Moreover, the retrieved snippets are often not grouped according to the APIs/libraries used, while they are only assessed for their functionality, disregarding their readability. In this work, we design a snippet mining methodology that receives queries in natural language and retrieves snippets, which are assessed not only for their functionality but also for their readability. The snippets are grouped according to their used API calls (libraries), thus enabling the developer to determine which solution is best fitted for his/her own source code, and making sure that it will be easily integrated and maintained. Upon providing a preliminary evaluation of our methodology on a set of different programming queries, we conclude that it can be effective in providing reusable and readable source code snippets.

Abstract. Lately, online issue tracking systems like Jira are used extensively for monitoring open-source software projects. Using these systems, different contributors can collaborate towards planning features and resolving issues that may arise during the software development process. In this context, several approaches have been proposed to extract knowledge from these systems in order to automate issue assignment. Though effective under certain scenarios, these approaches also have limitations; most of them are based mainly on textual features and they may use techniques that do not extract the underlying semantics and/or the expertise of the different contributors. Furthermore, they typically provide black-box recommendations, thus not helping the developers to interpret the issue assignments. In this work, we present an issue mining system that extracts semantic topics from issues and provides interpretable recommendations for issue assignments. Our system employs a dataset of Jira issues and extracts information not only from the textual features of issues but also from their components and their labels. These features, along with the extracted semantic topics, produce an aggregated model that outputs interpretable recommendations and useful statistics to support issue assignment. The results of our evaluation indicate that our system can be effective, leaving room for future research.

Abstract. Developers nowadays are increasingly turning to large language models (LLMs) like ChatGPT to assist them with coding tasks, inspired by the promise of efficiency and the advanced capabilities they offer. However, this raises important questions about the ease of integration and the safety of incorporating these tools into the development process. To investigate these questions, this paper examines a set of ChatGPT conversations. Upon annotating the conversations according to the intent of the developer, we focus on two critical aspects: firstly, the ease with which developers can produce suitable source code using ChatGPT, and, secondly, the quality aspects of the generated source code, determined by the compliance to standards and best practices. We research both the quality of the generated code itself and its impact on the project of the developer. Our results indicate that ChatGPT can be a useful tool for software development when used with discretion.

Abstract. Current state of practice dictates that software developers host their projects online and employ project management systems to monitor the development of product features, keep track of bugs, and prioritize task assignments. The data stored in these systems, if their semantics are extracted effectively, can be used to answer several interesting questions, such as finding who is the most suitable developer for a task, what the priority of a task should be, or even what is the actual workload of the software team. To support researchers and practitioners that work towards these directions, we have built a system that crawls data from the Jira management system, performs topic modeling on the data to extract useful semantics and stores them in a practical database schema. We have used our system to retrieve and analyze 656 projects of the Apache Software Foundation, comprising data from more than a million Jira issues.