Data Integration Strategy Research Articles

The role of national heritage in shaping Germany’s public health and data governance

Abstract Aim This paper aims to explain how political culture has influenced the scope of prevention measures, disease surveillance, and health data integration strategies in the German health system to date. Subject and methods Political culture is a major determinant of national health policies in countries, defining the means and scope of governmental authority for ensuring population health. This paper explains the role of political culture in shaping prevention and health promotion measures in the German health system, based on a public policy theory. Results During the post-war period, the structure of the German health system was (re-)designed to focus on curative medicine at the expense of public health. Current prevention and health promotion measures, often characterised as ‘too little, too late’, lead to medical treatments that are ‘too costly, too risky’. Linking data sources in Germany today is much more challenging than in other European countries, with health-relevant data often remaining in isolated silos that could be used for population health. Conclusion The analysis suggests that the national heritage shaping the political culture in Germany had a great influence on the limited role of government intervention, the interpretation of public health, and the state’s role in collecting and processing health data of citizens for research and policymaking.

A Novel intelligent SAV oriented QL-based task offloading in mobile edge environments

Edge computing is a novel and potential computing model which moves storage and computing capabilities to the network edge, substantially decreasing service latency and network traffic. The existing Internet of Things (IoT) network offloading algorithm faces limitations such as a fixed number of applications, high edge-to-edge delay, and reliance on a single Mobile Edge Computing (MEC) server, posing security and privacy concerns. Moreover, resource-constrained mobile devices need more effective data integration and compression strategies. Addressing these challenges, this study suggests an approach based on deep reinforcement learning (DRL), specifically the SAV (State Action Value) Oriented QL (Q-Learning) based Task Offloading method, to optimise task offloading and resource allocation in edge-cloud computing. The model aims to empower Mobile Devices (MDs) to develop optimal offloading decisions for long-term Quality Perception, utilising a neural network to establish the relationship between MD state and action value. The paper introduces a Recurrent Extended Memory Network (REMN) to capture dynamic workload behaviour at Edge nodes (ENs). It incorporates Quality Mapping, Quality Estimation, and a Quality-Aware DRL Task Offloading Algorithm to improve the accuracy and efficiency of the offloading procedure in MEC systems. This systematic approach improves overall system performance and enables MDs to leverage ENs for neural network training, reducing computational burdens. As a result, it can accomplish a more significant number of tasks, reducing latency from 0.74 ms to 7.168 ms and decreasing energy consumption from 270 J to 1820.39 J for tasks ranging from 10 to 50, respectively.

Expert opinions on improving coercion data collection across Europe: a concept mapping study.

Coercion is frequently used in mental health practice. Since it overrides some patients' fundamental human rights, adequate use of coercion requires legal and ethical justifications. Having internationally standardised datasets to benchmark and monitor coercion reduction programs is desirable. However, only a few countries have specific, open, publicly accessible registries for this issue. This study aims to assemble expert opinions regarding strategies that might be feasible for promoting, developing, and implementing an integrated and differentiated coercion data collection system in Europe at national and international levels. A concept mapping methodology was followed, involving 59 experts from 27 countries in generating, sorting and rating strategies regarding relevance and feasibility. The experts were all researchers and/or practitioner members of an EU-COST-Action focused on coercion reduction Fostering and Strengthening Approaches to Reducing Coercion in European Mental Health Services (FOSTREN). A hierarchical cluster analysis revealed a conceptual map of 41 strategies organized in seven clusters. These clusters fit into two higher-order domains: "Advancing Global Health Research: Collaboration, Accessibility, and Technological Innovations/Advancing International Research" and "Strategies for Comprehensive Healthcare Data Integration, Standardization, and Collaboration." Regarding the action with the higher priority, relevance was generally rated higher than feasibility. No differences could be found regarding the two domains regarding the relevance rating or feasibility of the respective strategies in those domains. The following strategies were rated as most relevant: "Collection of reliable data", "Implementation of nationwide register, including data on coercive measures", and "Equal understanding of different coercive measures". In analysing the differences in strategies between countries and their health prosperity, the overall rating did not differ substantially between the groups. The strategy rated as most relevant was the collection of reliable data in the nationwide health register, ensuring that countries share a standard understanding/definition of different coercive measures. Respondents did not consider the feasibility of establishing a shared European database for coercive measures to be high, nor did they envision the unification of mental health legislation in the future. There is some consensus on the most suitable strategies that can be adopted to enable international benchmarking of coercion in mental health settings.

Assessing the reproducibility of machine-learning-based biomarker discovery in Parkinson’s disease

Feature selection and machine learning algorithms can be used to analyze Single Nucleotide Polymorphisms (SNPs) data and identify potential disease biomarkers. Reproducibility of identified biomarkers is critical for them to be useful for clinical research; however, genotyping platforms and selection criteria for individuals to be genotyped affect the reproducibility of identified biomarkers. To assess biomarkers reproducibility, we collected five SNPs datasets from the database of Genotypes and Phenotypes (dbGaP) and explored several data integration strategies. While combining datasets can lead to a reduction in classification accuracy, it has the potential to improve the reproducibility of potential biomarkers. We evaluated the agreement among different strategies in terms of the SNPs that were identified as potential Parkinson’s disease (PD) biomarkers. Our findings indicate that, on average, 93% of the SNPs identified in a single dataset fail to be identified in other datasets. However, through dataset integration, this lack of replication is reduced to 62%. We discovered fifty SNPs that were identified at least twice, which could potentially serve as novel PD biomarkers. These SNPs are indirectly linked to PD in the literature but have not been directly associated with PD before. These findings open up new potential avenues of investigation.

Abstract 757: Integrating microbiome analysis and multi-modal data to identify high-risk population for esophageal adenocarcinoma

Abstract INTRODUCTION: Esophageal Adenocarcinoma (EAC) represents a pressing public health concern with a rising incidence and a staggering 5-year survival rate of less than 20%, emphasizing the critical need for early detection. Barrett's Esophagus (BE) serves as a well-established precursor to EAC, presenting an opportunity for proactive screening. The salivary microbiome provides a novel opportunity to identify a high-risk population with early-stage esophageal diseases. Here, we assessed whether the oral microbiome, in combination with lifestyle and clinical data, is associated with esophageal disease progression. METHODS: The study included n=23 esophageal adenocarcinoma patients, n=51 with Barrett's Esophagus, and n=32 healthy controls. Saliva samples were collected along with comprehensive lifestyle, dietary intake, and clinical data. Shotgun Metagenomics Sequencing was employed to investigate the microbiome composition of participants. Comprehensive microbiome analyses were performed to evaluate taxonomic profiles, including microbial abundance, microbiome diversity, and functional/metabolic insights. A multi-modal data integration strategy was used to incorporate data from the Diet Health Questionnaire (DHQ) and lifestyle and health records, aiming to utilize these multimodal features to predict esophageal disease formation and progression. RESULTS: A significant difference was observed in the salivary microbiome beta diversity between healthy controls and esophagus disease groups (p= 0.03). Clustering by microbiome taxonomic composition using the Dirichlet Multinomial Mixtures (DMM) framework indicates that some microbial signatures are associated with esophageal diseases (p=0.03). Healthy Eating Index Scores, total score (p=0.01); vegetables (p=0.003); greens, and beans score (p=0.009) were significantly different between study groups. Microbial pathways associated with diet, microbial taxa linked to lifestyle and medications, and supervised classification results reveal group characteristics and diversity variations based on key covariates such as B12 and sugar uptakes. CONCLUSION: The study presents an innovative analytic approach to identify high-risk populations for esophageal diseases by integrating microbiome analysis and multi-modal data, providing insights into early detection and prevention. The potential of this research lies in its ability to revolutionize screening strategies, offering a non-invasive and cost-effective means to identify individuals at risk of EAC at earlier, more treatable stages. Citation Format: Duygu Altinok Dindar, Madeline Krieger, Amy Palma, John Cheney, Fouad Otaki, Jessica Yu, Stephanie Wood, Thomas J. Sharpton, James McGann, Lisa Karstens, Gurkan G. Yardimci, Zhenzhen Zhang. Integrating microbiome analysis and multi-modal data to identify high-risk population for esophageal adenocarcinoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 757.

Holomics - a user-friendly R shiny application for multi-omics data integration and analysis

An organism’s observable traits, or phenotype, result from intricate interactions among genes, proteins, metabolites and the environment. External factors, such as associated microorganisms, along with biotic and abiotic stressors, can significantly impact this complex biological system, influencing processes like growth, development and productivity. A comprehensive analysis of the entire biological system and its interactions is thus crucial to identify key components that support adaptation to stressors and to discover biomarkers applicable in breeding programs or disease diagnostics. Since the genomics era, several other ’omics’ disciplines have emerged, and recent advances in high-throughput technologies have facilitated the generation of additional omics datasets. While traditionally analyzed individually, the last decade has seen an increase in multi-omics data integration and analysis strategies aimed at achieving a holistic understanding of interactions across different biological layers. Despite these advances, the analysis of multi-omics data is still challenging due to their scale, complexity, high dimensionality and multimodality. To address these challenges, a number of analytical tools and strategies have been developed, including clustering and differential equations, which require advanced knowledge in bioinformatics and statistics. Therefore, this study recognizes the need for user-friendly tools by introducing Holomics, an accessible and easy-to-use R shiny application with multi-omics functions tailored for scientists with limited bioinformatics knowledge. Holomics provides a well-defined workflow, starting with the upload and pre-filtering of single-omics data, which are then further refined by single-omics analysis focusing on key features. Subsequently, these reduced datasets are subjected to multi-omics analyses to unveil correlations between 2-n datasets. This paper concludes with a real-world case study where microbiomics, transcriptomics and metabolomics data from previous studies that elucidate factors associated with improved sugar beet storability are integrated using Holomics. The results are discussed in the context of the biological background, underscoring the importance of multi-omics insights. This example not only highlights the versatility of Holomics in handling different types of omics data, but also validates its consistency by reproducing findings from preceding single-omics studies.

Abstract A087: An ovarian cancer scRNA-seq atlas to dissect tumor-host interactions underlying metastatization and chemoresistance

Abstract The high tumor heterogeneity of high grade serous ovarian cancer (HGSOC) both intra-tumoral and inter-patient is one of the main factors hampering the identification of novel therapeutic treatment to improve survival of HGSOC patients. Single cell technologies are affording new inroads for the deep molecular characterization of such tumor heterogeneity, in both primary and metastatic samples. To this aim, we built a manually curated HGSOC transcriptomics atlas (~ 1,5 million cells, 79 patients) based on publicly available datasets and developed a data integration strategy specifically applicable to very heterogenous single cell data. scRNA-seq data integration is a central methodological challenge related to single cell data analysis, particularly in the context of highly heterogenous datasets. In this context, the main source of variability is the one existing among patients. Previously developed integration methods tend to perform badly on cancer data since they mainly rely on the assumption that cell populations present in different samples are similar, which might not apply for unknown, rare or patient-specific cell populations. Here we show a novel integration approach for patient-derived cancer data. To obtain a more faithful representation of the data, we performed identification of cell populations and metacells derivation separately for every sample to identify the distinct subpopulations characterizing each patient. In addition, to robustly represent the space describing the three main cell populations of the dataset, i.e., tumoral, immune and stromal, for each of the latter we selected the space identified by the union of highly variable genes, computed separately on every dataset, to preserve the variability of the system. Next, we fed a variational autoencoder with metacells data to integrate and characterize the different subpopulations that may be similar across samples. After integration, we were able to identify new robust gene signatures that would have been masked by datasets heterogeneity. The high degree of accuracy in the manual curation of metadata harmonization in the HGSOC atlas and the robustness and scalability of the data integration pipeline allowed us to obtain an in-depth characterization of the tumor subpopulations involved in therapy response by characterizing the chemotherapy induced transcriptional features. At the same time, thanks to the atlas-associated cell types labelling strategy, this atlas represents a transformative resource for the community by enabling the investigation of the interaction between tissue microenvironment (TME) cells and tumoral cells in shaping the metastatic process. This innovative HGSOC atlas will provide the scientific community of ovarian cancer research with a very powerful tool to investigate many aspects of this disease. The robust deep learning-based framework allows to expand the atlas easily and iteratively with newly generated datasets (out of sample extension), will constitute an invaluable resource for the HGSOC scientific field tackling HGSOC pathogenetic mechanism. Citation Format: Marta R. Sallese, Pietro Lo Riso, Carlo Emanuele Villa, Giuseppe Testa. An ovarian cancer scRNA-seq atlas to dissect tumor-host interactions underlying metastatization and chemoresistance [abstract]. In: Proceedings of the AACR Special Conference on Ovarian Cancer; 2023 Oct 5-7; Boston, Massachusetts. Philadelphia (PA): AACR; Cancer Res 2024;84(5 Suppl_2):Abstract nr A087.

Data integration strategies for whole-cell modeling.

Data makes the world go round-and high quality data is a prerequisite for precise models, especially for whole-cell models (WCM). Data for WCM must be reusable, contain information about the exact experimental background, and should-in its entirety-cover all relevant processes in the cell. Here, we review basic requirements to data for WCM and strategies how to combine them. As a species-specific resource, we introduce the Yeast Cell Model Data Base (YCMDB) to illustrate requirements and solutions. We discuss recent standards for data as well as for computational models including the modeling process as data to be reported. We outline strategies for constructions of WCM despite their inherent complexity.

Revolutionizing Data Management through Cloud-Based Data Fusion Platforms in Distributed Network Architectures

Data management is developing rapidly, and we need solutions that can handle massive volumes of diverse data. Especially for cloud-based data fusion and global network designs. Our research offers a fresh solution. Each difficult formula in this manner improves the system. Standardizing, matching, translating, and merging data from several sources is the fundamental strategy for data integration and management. We found that this alternative is superior to standard data management systems for growing, working fast, consistently, securely, and accurately integrating data, as well as cost-effectiveness. Data's visual presentation enhances the method's advantages and shows its potential. This research proves the technique works and illustrates how it may be utilized to advance the field. Supporting today's sophisticated data systems is a major advance. It's a solid, scalable data management solution that can evolve.

Comparative research on multi-source heterogeneous data fusion technologies

Both structured market data and unstructured financial news data can significantly affect stock price fluctuations. Therefore, relying only on a single data source for stock price trend prediction may produce information bias. This article combines these two types of heterogeneous data and systematically compares the effectiveness of different data fusion technologies to establish the optimal data integration strategy, thereby improving the accuracy of stock price prediction. Specifically, this paper innovatively proposes a stock price trend prediction model named MVL-SVM. The model successfully combines multi-perspective learning with support vector machine, and realizes the effective fusion of stock market data and financial news data. It is worth mentioning that the prediction accuracy of MVL-SVM model using direct fusion method is more than 30% higher than that of index modeling method, which indicates that direct fusion method can fully learn effective information from multiple information sources, thus significantly improving the prediction effect.

Cloud-based Data Analytics for Business Intelligence

Abstract: In the era of data-driven decision-making, organizations are turning to cloud-based data analytics for business intelligence to overcome the limitations of traditional on-premises systems. This paradigm shift offers the promise of scalable, agile, and advanced analytics capabilities. This paper explores the landscape of cloud- based data analytics for business intelligence by investigating existing systems, challenges, and opportunities. The study first examines leading cloud platforms such as Amazon Web Services (AWS) Redshift, Microsoft Azure Synapse Analytics, Google BigQuery, and Snowflake[11]. It evaluates their features, scalability, and integration options to assess their suitability for modern BI needs. Moreover, the research identifies critical challenges in transitioning to cloud-based analytics, including data integration complexities, security concerns, and cost management. The integration of advanced analytics techniques, such as machine learning and AI, into cloud- based environments is also explored. The study delves into the benefits and challenges of predictive analytics, anomaly detection, and other emerging capabilities that empower organizations to extract deeper insights from data. Furthermore, hybrid cloud architectures, which combine on-premises infrastructure with cloud resources, are investigated. Strategies for seamless data integration and workload distribution are discussed, enabling organizations to strike a balance between performance and data governance.

DeepAutoGlioma: a deep learning autoencoder-based multi-omics data integration and classification tools for glioma subtyping

Background and objectiveThe classification of glioma subtypes is essential for precision therapy. Due to the heterogeneity of gliomas, the subtype-specific molecular pattern can be captured by integrating and analyzing high-throughput omics data from different genomic layers. The development of a deep-learning framework enables the integration of multi-omics data to classify the glioma subtypes to support the clinical diagnosis.ResultsTranscriptome and methylome data of glioma patients were preprocessed, and differentially expressed features from both datasets were identified. Subsequently, a Cox regression analysis determined genes and CpGs associated with survival. Gene set enrichment analysis was carried out to examine the biological significance of the features. Further, we identified CpG and gene pairs by mapping them in the promoter region of corresponding genes. The methylation and gene expression levels of these CpGs and genes were embedded in a lower-dimensional space with an autoencoder. Next, ANN and CNN were used to classify subtypes using the latent features from embedding space. CNN performs better than ANN for subtyping lower-grade gliomas (LGG) and glioblastoma multiforme (GBM). The subtyping accuracy of CNN was 98.03% (± 0.06) and 94.07% (± 0.01) in LGG and GBM, respectively. The precision of the models was 97.67% in LGG and 90.40% in GBM. The model sensitivity was 96.96% in LGG and 91.18% in GBM. Additionally, we observed the superior performance of CNN with external datasets. The genes and CpGs pairs used to develop the model showed better performance than the random CpGs-gene pairs, preprocessed data, and single omics data.ConclusionsThe current study showed that a novel feature selection and data integration strategy led to the development of DeepAutoGlioma, an effective framework for diagnosing glioma subtypes.

Benchmarking strategies for cross-species integration of single-cell RNA sequencing data

The growing number of available single-cell gene expression datasets from different species creates opportunities to explore evolutionary relationships between cell types across species. Cross-species integration of single-cell RNA-sequencing data has been particularly informative in this context. However, in order to do so robustly it is essential to have rigorous benchmarking and appropriate guidelines to ensure that integration results truly reflect biology. Here, we benchmark 28 combinations of gene homology mapping methods and data integration algorithms in a variety of biological settings. We examine the capability of each strategy to perform species-mixing of known homologous cell types and to preserve biological heterogeneity using 9 established metrics. We also develop a new biology conservation metric to address the maintenance of cell type distinguishability. Overall, scANVI, scVI and SeuratV4 methods achieve a balance between species-mixing and biology conservation. For evolutionarily distant species, including in-paralogs is beneficial. SAMap outperforms when integrating whole-body atlases between species with challenging gene homology annotation. We provide our freely available cross-species integration and assessment pipeline to help analyse new data and develop new algorithms.

A novel approach to assess and improve syntactic interoperability in data integration

Data integration is essential to enrich a database with external information. One effective approach is to match shared identifiers across diverse databases. However, a lack of syntactic interoperability, which refers to the ability to match data based on their syntax, can pose challenges. In this paper, we present a novel method to evaluate and enhance syntactic interoperability, considering associated costs. First, we introduce the linking index and completeness index as generic measures of fine-grained syntactic interoperability. Second, we analyze the data consistency level of the identifiers using a rule-based framework for data quality assessment. Third, we propose a data integration strategy that strikes a balance between fixing data inconsistencies and the resulting benefits, as measured by the linking and completeness indices. The approach is illustrated through two use cases: bibliographic databases and clinical trial registries. The results demonstrate that standardizing identifiers’ representations can significantly improve syntactic interoperability in certain scenarios while in others, the standardization process does not yield improvements, discouraging, hence integration decisions. By conducting a cost–benefit analysis of improving data interoperability, this analysis enables data integrators to make informed decisions regarding the feasibility and advantages of proceeding with data integration.

Optimal Postprocessing Strategies With LSTM for Global Streamflow Prediction in Ungauged Basins

AbstractStreamflow prediction in ungauged basins (PUB) is challenging, and Long Short‐Term Memory (LSTM) is widely used to for such predictions, owing to its excellent migration performance. Traditional LSTM forced by meteorological data and catchment attribute data barely highlight the optimum data integration strategy for LSTM and its migration from data‐rich basins to ungauged ones. In this study, we experimented with 1,897 global catchments and found that LSTM‐corrected Global Hydrological Models (GHMs) outperformed uncorrected GHMs, improving the median Nash‐Sutcliff efficiency (NSE) from 0.03 to 0.66. Notably, there was a large gap between traditional LSTM modeling in ungauged basins and autoregressive modeling in data‐rich basins, and GHM‐forced LSTM were an effective way to close this gap in ungauged basins. The spatial heterogeneity of the performance of GHM‐forced LSTM was mainly influenced by three metrics (dryness, the leaf area index and latitude), which described the hydrological similarity among catchments. Weaker hydrological similarity among continental catchments results in larger variability in GHM‐forced LSTM, with the best performance in Siberia (NSE, 0.54) and the worst in North America (NSE, 0.10). However, the migration performance of GHM‐forced LSTM was significantly improved (NSE, 0.63) in ungauged basins when hydrological similarity was considered. This study stressed the advantages of GHM‐forced LSTM and due significance should be attached to hydrological similarities among catchments to improve hydrological prediction in ungauged catchments.

Advances in deep learning-based cancer outcome prediction using multi-omics data

Cancer prognosis reflects a complex biological process measured by multiple types of omics data. Deep learning frameworks have been proposed to integrate multi-omics data and predict patient outcomes in different cancer types, potentially revolutionizing cancer prognosis with superior performance. This minireview summarizes the advances in the strategies for multi-omics data integration and the performance of different deep learning models in prognosis prediction of diverse cancer types using multi-omics data published in the past 18 months. The challenges and limitations of deep learning models for predicting cancer outcomes based on multi-omics data are discussed.

Big data and genomics of Insulin to expand our knowledge

Physiology research must be guided moving forward by our significant investments represented by deposited data. We hypothesize that integrating these resources, namely our investments into genomics, makes it possible to discover many aspects of well-studied physiological pathways. Within this work, we build a genomics integration for the Insulin gene/protein, which serves as an example of data integrations that can be applied to any physiological system. The INS gene region has 2,076 unique variants from population genetics of hundreds of thousands of individuals (gnomAD, ClinVar, Geno2MP, Avada, dbSNP). Several variants are found near the transcriptional start site, enhancers, and following the INS transcripts that might influence the readthrough fusion-transcript INS- IGF2. This INS- IGF2 transcript splice site is confirmed within hundreds of pancreas RNAseq samples, lacks drift based on human genome sequencing, and has a possible elevated expression due to viral regulation within the pancreas/liver. Moreover, a rare, poorly characterized African-specific variant of INS-IGF2 results in a loss of the stop codon. INS transcript UTR variants rs689 and rs3842753, highly associated with Type 1 Diabetes, are found in hundreds of pancreas RNAseq datasets with an elevation of the 3’UTR alternatively spliced INS transcript. Finally, by combining literature, evolutionary profiling, and structural biology, we map rare missense variants involved in type-1 diabetes, type-2 diabetes, and monogenic/neonatal diabetes that influence Pre-Pro-Insulin translation, Pro-Insulin processing, dimer/hexamer secretory storage, receptor activation, and C-peptide detection for quasi-Insulin blood measurements. Finally, through these complex Insulin pathways and multiple animal model species, we highlight the many genes involved and how environmental factors such as cellular stress may interact with genetics to increase the prevalence of diabetes. Overall, our data integration strategy yields a platform for merging our investments into animal physiological model systems, human clinical genomics, and growing investments into genome sequencing to highlight the many exciting future research needs of physiological pathways. This research was funded by the Gerber Foundation (to JWP), National Institutes of Health (K01ES025435 to JWP), and Michigan State University. This is the full abstract presented at the American Physiology Summit 2023 meeting and is only available in HTML format. There are no additional versions or additional content available for this abstract. Physiology was not involved in the peer review process.

Omics data integration facilitates target selection for new antiparasitic drugs against TriTryp infections.

Introduction: Trypanosoma cruzi, Trypanosoma brucei, and Leishmania spp., commonly referred to as TriTryps, are a group of protozoan parasites that cause important human diseases affecting millions of people belonging to the most vulnerable populations worldwide. Current treatments have limited efficiencies and can cause serious side effects, so there is an urgent need to develop new control strategies. Presently, the identification and prioritization of appropriate targets can be aided by integrative genomic and computational approaches. Methods: In this work, we conducted a genome-wide multidimensional data integration strategy to prioritize drug targets. We included genomic, transcriptomic, metabolic, and protein structural data sources, to delineate candidate proteins with relevant features for target selection in drug development. Results and Discussion: Our final ranked list includes proteins shared by TriTryps and covers a range of biological functions including essential proteins for parasite survival or growth, oxidative stress-related enzymes, virulence factors, and proteins that are exclusive to these parasites. Our strategy found previously described candidates, which validates our approach as well as new proteins that can be attractive targets to consider during the initial steps of drug discovery.

Machine Learning-Assisted Approaches in Modernized Plant Breeding Programs

In the face of a growing global population, plant breeding is being used as a sustainable tool for increasing food security. A wide range of high-throughput omics technologies have been developed and used in plant breeding to accelerate crop improvement and develop new varieties with higher yield performance and greater resilience to climate changes, pests, and diseases. With the use of these new advanced technologies, large amounts of data have been generated on the genetic architecture of plants, which can be exploited for manipulating the key characteristics of plants that are important for crop improvement. Therefore, plant breeders have relied on high-performance computing, bioinformatics tools, and artificial intelligence (AI), such as machine-learning (ML) methods, to efficiently analyze this vast amount of complex data. The use of bigdata coupled with ML in plant breeding has the potential to revolutionize the field and increase food security. In this review, some of the challenges of this method along with some of the opportunities it can create will be discussed. In particular, we provide information about the basis of bigdata, AI, ML, and their related sub-groups. In addition, the bases and functions of some learning algorithms that are commonly used in plant breeding, three common data integration strategies for the better integration of different breeding datasets using appropriate learning algorithms, and future prospects for the application of novel algorithms in plant breeding will be discussed. The use of ML algorithms in plant breeding will equip breeders with efficient and effective tools to accelerate the development of new plant varieties and improve the efficiency of the breeding process, which are important for tackling some of the challenges facing agriculture in the era of climate change.

Domain adaptation for supervised integration of scRNA-seq data

Large-scale scRNA-seq studies typically generate data in batches, which often induce nontrivial batch effects that need to be corrected. Given the global efforts for building cell atlases and the increasing number of annotated scRNA-seq datasets accumulated, we propose a supervised strategy for scRNA-seq data integration called SIDA (Supervised Integration using Domain Adaptation), which uses the cell type annotations to guide the integration of diverse batches. The supervised strategy is based on domain adaptation that was initially proposed in the computer vision field. We demonstrate that SIDA is able to generate comprehensive reference datasets that lead to improved accuracy in automated cell type mapping analyses.