Type 2 diabetes (T2D) is a common chronic disease whose etiology is known to have a strong genetic component. Standard genetic approaches, although allowing for the detection of a number of gene variants associated with the disease as well as differentially expressed genes, cannot fully explain the hereditary factor in T2D. The explosive growth in the genomic sequencing technologies over the last decades provided an exceptional impetus for transcriptomic studies and new approaches to gene expression measurement, such as RNA-sequencing (RNA-seq) and single-cell technologies. The transcriptomic analysis has the potential to find new biomarkers to identify risk groups for developing T2D and its microvascular and macrovascular complications, which will significantly affect the strategies for early diagnosis, treatment, and preventing the development of complications. In this article, we focused on transcriptomic studies conducted using expression arrays, RNA-seq, and single-cell sequencing to highlight recent findings related to T2D and challenges associated with transcriptome experiments.

In recent years, great advances have been made in the field of collection, storage, and analysis of biological samples. Large collections of samples, biobanks, have been established in many countries. Biobanks typically collect large amounts of biological samples and associated clinical information; the largest collections include over a million samples. In this review, we summarize the main directions in which biobanks aid medical genetics and genomic research, from providing reference allele frequency information to allowing large-scale cross-ancestry meta-analyses. The largest biobanks greatly vary in the size of the collection, and the amount of available phenotype and genotype data. Nevertheless, all of them are extensively used in genomics, providing a rich resource for genome-wide association analysis, genetic epidemiology, and statistical research into the structure, function, and evolution of the human genome. Recently, multiple research efforts were based on trans-biobank data integration, which increases sample size and allows for the identification of robust genetic associations. We provide prominent examples of such data integration and discuss important caveats which have to be taken into account in trans-biobank research.

Complications endangering mother or fetus affect around one in seven pregnant women. Investigation of the genetic susceptibility to such diseases is of high importance for better understanding of the disease biology as well as for prediction of individual risk. In this study, we collected and analyzed GWAS summary statistics from the FinnGen cohort and UK Biobank for 24 pregnancy complications. In FinnGen, we identified 11 loci associated with pregnancy hypertension, excessive vomiting, and gestational diabetes. When UK Biobank and FinnGen data were combined, we discovered six loci reaching genome-wide significance in the meta-analysis. These include rs35954793 in FGF5 ($p=6.1\times {10}^{-9}$), rs10882398 in PLCE1 ($p=8.9\times {10}^{-9}$), and rs167479 in RGL3 ($p=5.2\times {10}^{-9}$) for pregnancy hypertension, rs10830963 in MTNR1B ($p=4.5\times {10}^{-41}$) and rs36090025 in TCF7L2 ($p=3.4\times {10}^{-15}$) for gestational diabetes, and rs2963457 in the EBF1 locus ($p=6.5\times {10}^{-9}$) for preterm birth. In addition to the identified genome-wide associations, we also replicated 14 out of 40 previously reported GWAS markers for pregnancy complications, including four more preeclampsia-related variants. Finally, annotation of the GWAS results identified a causal relationship between gene expression in the cervix and gestational hypertension, as well as both known and previously uncharacterized genetic correlations between pregnancy complications and other traits. These results suggest new prospects for research into the etiology and pathogenesis of pregnancy complications, as well as early risk prediction for these disorders.

Although traditional statistical techniques and machine learning methods have contributed significantly to genetics and, in particular, inherited disease diagnosis, they often struggle with complex, high-dimensional data, a challenge now addressed by state-of-the-art deep learning models. Large language models (LLMs), based on transformer architectures, have excelled in tasks requiring contextual comprehension of unstructured medical data. This systematic review examines the role of generative Artificial Intelligence (AI) methods in human medical genomics, focusing on the genetic research and diagnostics of both rare and common diseases. Automated keyword-based search in PubMed, bioRxiv, medRxiv, and arXiv was conducted, targeting studies on LLM applications in diagnostics and education within genetics and removing irrelevant or outdated models. A total of 195 studies were analyzed, highlighting the prospects of their applications in knowledge navigation, analysis of clinical and genetic data, and interaction with patients and medical professionals. Key findings indicate that while transformer-based models perform well across a diverse range of tasks (such as identification of tentative molecular diagnosis from clinical data or genetic variant interpretation), major challenges persist in integrating multimodal data (genomic sequences, imaging, and clinical records) into unified and clinically robust pipelines, facing limitations in generalizability and practical implementation in clinical settings. This review provides a comprehensive classification and assessment of the current capabilities and limitations of LLMs in transforming hereditary disease diagnostics and supporting genetic education, serving as a guide to navigate this rapidly evolving field, while outlining application use cases, implementation guidance, and forward-looking research directions.

Guar gum, a polysaccharide derived from guar seeds, is widely used in a variety of industrial applications, including oil and gas production. Although guar is mostly propagated in India, interest in guar as a new industrial legume crop is increasing worldwide, demanding the development of effective tools for marker-assisted selection. In this paper, we report a wide-ranging set of 4907 common SNPs and 327 InDels generated from RADseq genotyping data of 166 guar plants of different geographical origin. A custom guar reference genome was assembled and used for variant calling. A consensus set of variants was built using three bioinformatic pipelines for short variant discovery. The developed molecular markers were used for genome-wide association study, resulting in the discovery of six markers linked to the variation of an important agronomic trait—percentage of pods matured to the harvest date under long light day conditions. One of the associated variants was found inside the putative transcript sequence homologous to an ABC transporter in Arabidopsis, which has been shown to play an important role in D-myo-inositol phosphates metabolism. Earlier, we suggested that genes involved in myo-inositol phosphate metabolism have significant impact on the early flowering of guar plants. Hence, we believe that the developed SNP set allows for the identification of confident molecular markers of important agrobiological traits.

The COVID-19 pandemic has drawn the attention of many researchers to the interaction between pathogen and host genomes. Over the last two years, numerous studies have been conducted to identify the genetic risk factors that predict COVID-19 severity and outcome. However, such an analysis might be complicated in cohorts of limited size and/or in case of limited breadth of genome coverage. In this work, we tried to circumvent these challenges by searching for candidate genes and genetic variants associated with a variety of quantitative and binary traits in a cohort of 840 COVID-19 patients from Russia. While we found no gene- or pathway-level associations with the disease severity and outcome, we discovered eleven independent candidate loci associated with quantitative traits in COVID-19 patients. Out of these, the most significant associations correspond to rs1651553 in MYH14p = 1.4 × 10⁻⁷), rs11243705 in SETX (p = 8.2 × 10⁻⁶), and rs16885 in ATXN1 (p = 1.3 × 10⁻⁵). One of the identified variants, rs33985936 in SCN11A, was successfully replicated in an independent study, and three of the variants were found to be associated with blood-related quantitative traits according to the UK Biobank data (rs33985936 in SCN11A, rs16885 in ATXN1, and rs4747194 in CDH23). Moreover, we show that a risk score based on these variants can predict the severity and outcome of hospitalization in our cohort of patients. Given these findings, we believe that our work may serve as proof-of-concept study demonstrating the utility of quantitative traits and extensive phenotyping for identification of genetic risk factors of severe COVID-19.