Biology subjects

Frosi, G.

Publications and source records attributed to Frosi, G..

3 recordsLinked to original sources

bioRxiv2026

A multimodal characterization of the human uncinate fasciculus

The uncinate fasciculus (UF) is a hook-shaped long-range association white matter tract that serves to bidirectionally transmit information between the anterior temporal lobe and the orbitofrontal cortex. Neuroimaging studies have suggested that changes in UF microstructure are involved in the neurobiological sequalae of childhood abuse (CA). Given that the UF is not present in rodents, it is vastly understudied with no cellular and molecular information available. To this end, we aimed to perform a multimodal characterization of the UF between individuals diagnosed with depression who died by suicide with (DS-CA) and without a history of severe CA (DS) and psychiatrically healthy individuals (CTRL). Fresh frozen UF tissue was obtained from the Douglas Bell-Canada Brain Bank, with phenotypic information collected via psychological autopsy. Immunohistochemistry with PDGFR and NogoA was used to label oligodendrocyte precursor cell (OPC) and oligodendrocyte (OL), respectively, and stereology was performed to ascertain cell density and soma volume. Single nucleus RNA sequencing (snRNAseq) was used to generate a transcriptomic survey of the cell types found in the UF. Finally, spectral focusing Coherent Anti-Stokes Raman Scattering (sf-CARS) microscopy was employed in tandem with a custom AxonDeepSeg segmentation model to measure axon diameter, myelin thickness, and g-ratio. No group differences were observed in histology or ultrastructure metrics, but nearly 50 differentially expressed genes (DEG) were identified between groups. Interestingly NECTIN3, the top DEG downregulated in OL1 and OL3 of DS-CA, is a computationally predicted target of the microRNA MIR646, the host gene of which was significantly downregulated in multiple cell types in DS-CA. Age-associated changes were pronounced and observed in all modalities, including an age-related increase in OL density, extensive changes in glial gene expression, as well as decreases in axon diameter and g-ratio. This study serves as a foundational resource on the molecular and cellular properties of the human UF. Our results suggest that observable myelin-related traces of depression or CA are limited in the UF and highlights the need for future research on the cellular and molecular properties of white matter tracts during aging.

neuroscience↗

bioRxiv2026

A glucocorticoid-responsive polygenic signature in the anterior cingulate cortex moderates the association of early-life adversity and vulnerability for depression

Stress exposure is a major risk factor for psychopathology, yet how stress mediators shape long-term psychiatric vulnerability in humans remains unclear. Glucocorticoids, central effectors of the stress response, regulate gene expression through tissue-specific transcriptional programs, suggesting that glucocorticoid-responsive networks may shape sensitivity to adversity. Using RNA-sequencing following chronic glucocorticoid exposure in a non-human primate model, we identified a gene co-expression network specific to the anterior cingulate cortex (ACC) that was highly preserved across human post-mortem brain datasets relevant to depression. We derived an expression-based polygenic score (ePGS) reflecting genetic variation in network activity and tested its interaction with adversity in the UK Biobank. The ACC-specific glucocorticoid-responsive ePGS moderated the association between adversity and depressive symptoms in adult females, with the strongest effects for early-life adversity. Network genes were enriched for neurodevelopmental processes and showed stronger co-expression during childhood, highlighting a developmentally sensitive, region-specific mechanism linking stress exposure to depression risk.

neuroscience↗

bioRxiv2025

Leveraging the largest harmonized epigenomic data collection for metadata prediction validated and augmented over 350,000 public epigenomic datasets

Epigenomic data found in public databases often suffer from issues of non-standardization and incompleteness in their associated metadata. There are currently no automated approaches to validate or correct missing or inaccurate information listed in databases. To tackle this challenge, we harnessed the extensive harmonized data and metadata provided by the EpiATLAS project of the International Human Epigenome Consortium (IHEC) to train EpiClass, a suite of machine learning classifiers that can predict key metadata ([~]98% accuracy), including experimental assay, donor sex, biospecimen and sample cancer status. The development of these classifiers enabled the identification of a few mislabeled and low-quality datasets in the EpiATLAS project, while also completing with high-confidence most of the missing metadata. These classifiers were also validated on ENCODE datasets absent from the initial training, then applied to assess more than 350,000 human ChIP-Seq and RNA-Seq datasets from public repositories. Overall, this effort not only validated the accuracy of the vast majority of assays reported by the original authors, but also unveiled [~]500 datasets with discrepancies, in particular through data swap within series of experiments. More importantly, EpiClass also supplied high-confidence predictions for over 320,000 metadata attributes of the biological sample such as the sex, cancer status and biomaterial type, which had been originally omitted in the majority of cases. Our work introduces the first systematic approach for metadata correction and augmentation, enhancing the quality and reliability of publicly available epigenomic data.

bioinformatics↗