DZDiana Zagirova

Diana Zagirova

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Diana Zagirova has a total of 31 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Accelerated Aging Signatures in 3D Genome Organization and Transcriptome in Schizophrenia

    by K.A. Ulianov, Diana R Zagirova, Anna D Kononkova, Anastasiia V Dudkovskaia, Maria N Molodova, Kirill V Morozov, O.I. Efimova, Maria Bazarevich, Alexander V Cherkasov, Polina D Morozova, et al.

    Schizophrenia is a severe neuropsychiatric disorder that affects the behavioral, emotional and cognitive state of patients. Despite its substantial heritability, the molecular etiology of the disease remains poorly understood. Many schizophrenia-associated genetic variants reside in non-coding regions, and exert their effects through distal regulatory elements of the genome. In this context, the three-dimensional organization of the genome is expected to play a decisive role in establishing contacts between these regulatory elements and their target genes, thereby mediating schizophrenia-associated dysregulation of gene expression. Here, we present a novel Hi-C dataset providing an unprecedented view of three-dimensional genome organization in post-mortem schizophrenia brain samples. Our findings indicate that most changes occur at long-range genomic distances while local architecture of topologically-associated domains remains largely intact. However, neurons display localized and functionally relevant loop differences, particularly in regulatory regions associated with neurodevelopmental processes. Global characteristics of higher-order chromatin organization show accelerated aging alteration pattern in schizophrenia, and downstream analysis of transcriptomic data in schizophrenia brain samples further confirms that schizophrenia is associated with accelerated aging.

  • Open Access

    The End of Aging Clocks: Training Foundation Models to Reason in Aging and Longevity

    by Alex Zhavoronkov, Vladimir Aladinskyi, Alex Aliper, Zulfat Miftakhutdinov, Mathieu Reymond, Vladimir Naumov, Diana Zagirova, Stefan Pushkov, Denis Sidorenko, Rim Shayakhmetov, et al.

    The aging clock paradigm has yielded dozens of specialist models that can estimate chronological age or mortality from virtually any biodata type. Yet each such model operates within a fixed modality, relies on a predetermined feature set, and produces limited biological interpretation. Here, we report Longevity-LLM v0.1, a Qwen3-14B model fine-tuned through supervised and reinforcement learning regimes on DNA methylation, proteomics, clinical biomarker, and RNA expression data. Longevity-LLM achieves high ranks in the recently announced Longevity Bench, including such tasks as cancer survival and RNA- or proteome-based age prediction. After reinforcement fine-tuning, the model achieved a 4.34-year MAE in epigenetic age prediction, surpassing the Horvath multi-tissue clock. In addition to age prediction, Longevity-LLM can carry out numerous other tasks, including proteomic profile generation, for which it significantly outperforms all frontier LLMs. These results demonstrate that a single modestly sized LLM can match or replace purpose-built aging clocks across data modalities. This work constitutes an interim report from the initial sprint of our Multi-Modal AI Gym for Science (MMAI), an initiative dedicated to building foundation models for drug discovery and aging research.

  • Open Access

    LongevityBench: Are SotA LLMs ready for aging research?

    by Alex Zhavoronkov, Denis Sidorenko, Vladimir Naumov, Stefan Pushkov, Diana Zagirova, Vladimir Aladinskiy, Derya Unutmaz, Alex Aliper, Fedor Galkin

    Aging is a core biological process observed in most species and tissues, which is studied with a vast array of technologies. We argue that the abilities of AI systems to emulate aging and to accurately interpret biodata in its context are the key criteria to judge an LLM’s utility in biomedical research. Here, we present LongevityBench — a collection of tasks designed to assess whether foundation models grasp the fundamental principles of aging biology and can use low-level biodata to arrive at phenotype-level conclusions. The benchmark covers a variety of prediction targets including human time-to-death, mutations’ effect on lifespan, and age-dependent omics patterns. It spans all common biodata types used in longevity research: transcriptomes, DNA methylation profiles, proteomes, genomes, clinical blood tests and biometrics, as well as natural language annotations. After ranking state-of-the-art foundation models using LongevityBench, we highlight their weaknesses and outline procedures to maximize their utility in aging research and life sciences.

  • Open Access

    Fetal signatures in the 3D genome of iPSC-derived neurons and their implications for disease modeling

    by D.R. Zagirova, A.D. Kononkova, K.V. Morozov, M.N. Molodova, N.S. Vaulin, A.V. Dudkovskaia, P.I. Dozorova, O.I. Efimova, A.V. Tvorogova, K.A. Ulianov, et al.

    Induced pluripotent stem cells (iPSCs) have revolutionized neuroscience, providing an approach to generate patient-specific neurons for modeling of neurological diseases. However, it remains unclear how closely iPSC-derived neurons replicate the chromatin architecture of authentic brain neurons. Here, we uniformly process datasets for 228 human and 89 mouse Hi-C and Snm3C-seq samples of different cell subtypes merged into 96 high-coverage contact maps used to examine chromatin features ranging from chromatin compartments and topologically associating domains (TADs) to chromatin loops, Polycomb-mediated contacts, and frequently interacting regions (FIREs). We find that iPSC-derived neurons largely retain chromatin state of undifferentiated cells and resemble fetal rather than mature neurons. iPSC-derived neurons exhibit unusually strong compartmentalization, an enrichment of developmental genes at TAD borders, and a marked reduction of long-range repressive Polycomb-mediated contacts that typically silence early fetal programs. Although immature, iPSC-derived neurons offer advantages for modeling interactions between disease-associated SNPs and target genes, as many psychiatric disorders have neurodevelopmental origins. Integrating iPSC-derived and postmortem neuronal datasets therefore provides complementary insights into the chromatin landscape underlying disease-associated interactions. Our study offers a valuable Hi-C resource for the community and provides a detailed comparison of chromatin architecture throughout neuronal maturation, underscoring its importance for validating neuronal models and providing a robust framework for future studies.