Biology subjects

Alentorn, A.

Publications and source records attributed to Alentorn, A..

6 recordsLinked to original sources

bioRxiv2026

VelOT: kinetic-free RNA velocity inference via optimal transport, flow-field smoothing, and VAMP coarse-graining of cellular dynamics

Inferring cellular dynamics from snapshot single-cell RNA sequencing remains difficult when spliced and unspliced counts are sparse or unreliable. We present VelOT, a kinetic-free RNA velocity framework that formulates dynamics as local optimal transport on the gene-expression manifold. VelOT orders cells by diffusion pseudotime, constructs overlapping spatial-temporal windows, estimates displacement vectors with entropy-regularized transport, and smooths them with a lightweight neural flow field. A downstream VAMP-based MetaFlow module learns soft meta-states and a directed PAGA-like graph, identifying initial, terminal, branching, and cycling regimes with committor probabilities. Across four real benchmarks and three synthetic topologies, VelOT outperforms scVelo, DeepVelo, and FluxMatching in cross-boundary directionality and intra-cluster coherence while remaining computationally efficient. In adult oligodendroglioma scRNA-seq, VelOT recovers stem-like to astrocyte-like and oligodendrocyte-like differentiation axes without kinetic inputs. VelOT reframes RNA velocity within scRNA-seq as a geometry and transport problem that does not require kinetic modeling.

bioinformatics↗

bioRxiv2026

h5adify: neuro-symbolic metadata harmonizationenables scalable AnnData integration with locallarge language models

BackgroundThe rapid growth of public single-cell and spatial transcriptomics repositories has shifted the main bottleneck for atlas-scale integration from data generation to metadata heterogeneity. Even when datasets are released in the AnnData H5AD format, inconsistent column naming, partial annotations, and mixed gene identifier conventions frequently prevent reproducible merging, downstream benchmarking, and reuse in foundation model training. Automated approaches that resolve semantic inconsistency while preserving biological validity are therefore essential for scalable data reuse. ResultsWe present h5adify, a neuro-symbolic toolkit that combines deterministic biological inference with locally deployed large language models to transform heterogeneous AnnData objects into schema-normalized, integration-ready representations. The framework performs metadata field discovery, gene identifier harmonization, optional paper-aware extraction, and consensus resolution with explicit uncertainty logging. Benchmarking four open-weight model families deployed through Ollama (Gemma, Llama, Mistral, and Qwen) demonstrates that small local models achieve high semantic accuracy in metadata resolution with low hallucination rates and modest computational requirements. In controlled simulations introducing annotation noise into single-cell and Visium-like datasets, harmonization improves integration benchmarking and reduces spurious batch effects. Application to sex-annotated glioblastoma datasets recovers biologically coherent microenvironmental patterns and cell type-specific genomic differences not explained by differential expression alone. ConclusionsTogether, h5adify provides a reproducible framework for evaluating LLM-assisted biocuration and enables scalable, privacy-preserving metadata harmonization for modern single-cell atlases and foundation model pipelines. These results demonstrate that modular neuro-symbolic integration of deterministic biological inference and small local language models can effectively resolve semantic heterogeneity while remaining computationally accessible.

bioinformatics↗

bioRxiv2024

The oncometabolite D-2-hydroxyglutarate promotes DNA hypermethylation at lineage-specific enhancers controlling microglial activation in IDHmut gliomas

Tumor-associated microglia and macrophages (TAMs), the most abundant myeloid populations in gliomas, shape immune responses through transcriptional programs influenced by the tumor microenvironment. Although these programs differ according to tumor IDH status, the underlying epigenetic mechanisms remain poorly understood. Here, we uncover widespread DNA hypermethylation in the myeloid compartment of IDH-mutant gliomas, predominantly at distal enhancers enriched for motifs of core microglial transcription factors (TFs). This remodeled enhancer landscape strongly correlated with reduced activity of TF regulons and coordinated repression of immunomodulatory programs that normally support microglial activation. Using primary human microglia, we show that prolonged exposure to the oncometabolite D-2-hydroxyglutarate (D-2HG) reduces TET activity and increases 5mC/5hmC ratios near TF-binding motifs within enhancers affected ex vivo. Consistent with these epigenetic alterations, D-2HG-treated microglia exhibited transcriptional signatures compatible with blunted proinflammatory responses, whereas pharmacological inhibition of mutant IDH in patients partially restored microglial immune reactivity. Altogether, our findings reveal a chronic D-2HG-driven epigenetic priming mechanism that promotes a hyporesponsive microglial state, providing a rationale for the immunologically cold phenotype of IDH-mutant gliomas and offering insight into how IDH-targeted therapies may reshape microglial immune responses.

cancer biology↗

bioRxiv2022

Cellular senescence in malignant cells promotes tumor progression in mouse and patient Glioblastoma

Glioblastoma (GBM) is the most common primary malignant brain tumor in adults, yet it remains refractory to systemic therapy. Elimination of senescent cells has emerged as a promising new treatment approach against cancer. Here, we investigated the contribution of senescent cells to GBM progression. Senescent cells were identified in patient and mouse GBMs. Partial removal of p16Ink4a-expressing malignant senescent cells, which make up less than 7 % of the tumor, modified the tumor ecosystem and improved the survival of GBM-bearing mice. By combining single cell and bulk RNA sequencing, immunohistochemistry and genetic knockdowns, we identified the NRF2 transcription factor as a determinant of the senescent phenotype. Remarkably, our mouse senescent transcriptional signature and underlying mechanisms of senescence are conserved in patient GBMs, in whom higher senescence scores correlate with shorter survival times. These findings suggest that senolytic drug therapy may be a beneficial adjuvant therapy for patients with GBM.

cancer biology↗

bioRxiv2021

Pan-cancer landscape of AID-related mutations, composite mutations and its potential role in the ICI response

Activation-induced cytidine deaminase, AICDA or AID, is a driver of somatic hypermutation and class-switch recombination in immunoglobulins. In addition, this deaminase belonging to the APOBEC family, may have off-target effects genome-wide, but its effects at pan-cancer level are not well elucidated. Here, we used different pan-cancer datasets, totaling more than 50,000 samples analyzed by whole-genome, whole-exome or targeted sequencing. AID synergizes initial hotspot mutations by a second composite mutation. Analysis of 2.5 million cells, normal and oncogenic, revealed AICDA expression activation after oncogenic transformation and cell cycle regulation loss. AID mutational load was found to be independently associated with favorable outcome in immune-checkpoint inhibitors (ICI) treated patients across cancers after analyzing 2,000 samples. Finally, we found that AID related neoepitopes, resulting from mutations at more frequent hotspots if compared to other mutational signatures, enhance CXCL13/CCR5 expression, immunogenicity and T-cell exhaustion, which may increase ICI sensitivity. In BriefA combined bulk and single cell multi-omic analysis of over 50,000 patients and 2.5 million cells across 80 tumor types reveals oncogenic acquired AICDA expression inducing composite mutations and clonal immunogenic neoepitopes that are associated with favorable outcome in patients treated by immune-checkpoint inhibitors. Highlights* Pan-cancer analysis of AID mutations using > 50,000 samples, 2,000 ICI treated cases and 2.5 million cells with genome, exome and transcriptome data * Oncogenic transient AICDA expression induces mutations mainly during transcription of its off-target genes in virtually all cancers * AID is implicated in composite mutations on weakly functional alleles and immunogenic clonal neoepitopes at hotspots with greater positive selection * AID mutational load predicts response and is associated with favorable outcome in ICI treated patients

bioinformatics↗

bioRxiv2020

Learning to leverage salient regions in neuro-oncology using Deap Learning

Developing an automatic tumor detector for MRI medical images is a major challenge in neuro-oncology. The availability of such a tool would be a valuable assistance for the radiologists. Numerous works have tried to segment the tumor tissues, others have attempted to localize the tumor globally. In this work we focus on this second class of methods and we compare two drastically different strategies. The first one is an assumption-free anomaly detector build over a Variational Auto-Encoder (VAE), and the second one is a VGG classifier that embed Attention-Gated (AG) units to focus on the target structures at almost no additional computational cost. This comparison is first conducted on the publicly available BraTS glioma dataset for which published performance results can serve as reference, and extended as such (ie., without transfer learning) to two internal image datasets, namely Primary Central Nervous System Lymphoma (PCNSL) and Metastasis. The results demonstrate that the VAE and AG-VGG strategies can be used, up to a certain extent, to localize brain tumors.

cancer biology↗