Evaluation of the Patients with the Diagnosis of Pontocerebellar Hypoplasia: A Multicenter National Study.

Pontocerebellar hypoplasia (PCH) is a heterogeneous group of neurodegenerative disorders characterized by hypoplasia and degeneration of the cerebellum and pons. We aimed to identify the clinical, laboratory, and imaging findings of the patients with diagnosed PCH with confirmed genetic analysis. We collected available clinical data, laboratory, and imaging findings in our retrospective multicenter national study of 64 patients with PCH in Turkey. The genetic analysis included the whole-exome sequencing (WES), targeted next-generation sequencing (NGS), or single gene analysis. Sixty-four patients with PCH were 28 female (43.8%) and 36 (56.3%) male. The patients revealed homozygous mutation in 89.1%, consanguinity in 79.7%, pregnancy at term in 85.2%, microcephaly in 91.3%, psychomotor retardation in 98.4%, abnormal neurological findings in 100%, seizure in 63.8%, normal biochemistry and metabolic investigations in 92.2%, and dysmorphic findings in 51.2%. The missense mutation was found to be the most common variant type in all patients with PCH. It was detected as CLP1 (n = 17) was the most common PCH related gene. The homozygous missense variant c.419G > A (p.Arg140His) was identified in all patients with CLP1. Moreover, all patients showed the same homozygous missense variant c.919G > T (p.A307S) in TSEN54 group (n = 6). In Turkey, CLP1 was identified as the most common causative gene with the identical variant c.419G > A; p.Arg140His. The current study supports that genotype data on PCH leads to phenotypic variability over a wide phenotypic spectrum.

A microdeletion event at 19q13.43 in IDH-mutant astrocytomas is strongly correlated with MYC overexpression.

MYC dysregulation is pivotal in the onset and progression of IDH-mutant gliomas, mostly driven by copy-number alterations, regulatory element alterations, or epigenetic changes. Our pilot analysis uncovered instances of relative MYC overexpression without alterations in the proximal MYC network (PMN), prompting a deeper investigation into potential novel oncogenic mechanisms. Analysing comprehensive genomics profiles of 236 "IDH-mutant 1p/19q non-co-deleted" lower-grade gliomas from The Cancer Genome Atlas, we identified somatic genomic alterations within the PMN. In tumours without PMN-alterations but with MYC-overexpression, genes correlated with MYC-overexpression were identified. Our analyses yielded that 86/236 of astrocytomas exhibited no PMN-alterations, a subset of 21/86 displaying relative MYC overexpression. Within this subset, we discovered 42 genes inversely correlated with relative MYC expression, all on 19q. Further analysis pinpointed a minimal common region at 19q13.43, encompassing 15 genes. The inverse correlations of these 15 genes with relative MYC overexpression were re-confirmed using independent scRNAseq data. Further, the micro-deleted astrocytoma subset displayed significantly higher genomic instability compared to WT cases, but lower instability compared to PMN-hit cases. This newly identified 19q micro-deletion represents a potential novel mechanism underlying MYC dysregulation in astrocytomas. Given the prominence of 19q loss in IDH-mutant gliomas, our findings bear significant implications for understanding gliomagenesis.

StemnesScoRe: an R package to estimate the stemness of glioma cancer cells at single-cell resolution.

Background/aim: Glioblastoma is the most heterogeneous and the most difficult-to-treat type of brain tumor and one of the deadliest among all cancers. The high plasticity of glioma cancer stem cells and the resistance they develop against multiple modalities of therapy, along with their high heterogeneity, are the main challenges faced during treatment of glioblastoma. Therefore, a better understanding of the stemness characteristics of glioblastoma cells is needed. With the development of various single-cell technologies and increasing applications of machine learning, indices based on transcriptomic and/or epigenomic data have been developed to quantitatively measure cellular states and stemness. In this study, we aimed to develop a glioma-specific stemness score model using scATAC-seq data for the first time. Materials and methods: We first applied three powerful machine-learning algorithms, i.e. random forest, gradient boosting, and extreme gradient boosting, to glioblastoma scRNA-seq data to discover the most important genes associated with cellular states. We then identified promoter and enhancer regions associated with these genes. After downloading the scATAC-seq peaks and their read counts for each patient, we identified the overlapping regions between the single-cell peaks and the peaks of genes obtained through machine-learning algorithms. Then we calculated read counts that were mapped to these overlapping regions. We finally developed a model capable of estimating the stemness score for each glioma cell using overlapping regions and the importance of genes predictive of glioblastoma cellular states. We also created an R package, accessible to all researchers regardless of their coding proficiency. Results: Our results showed that mesenchymal-like stem cells display higher stemness scores compared to neural-progenitor-, oligodendrocyte-progenitor-, and astrocyte-like cells. Conclusion: scATAC-seq can be used to assess heterogeneity in glioblastoma and identify cells with high stemness characteristics. The package is publicly available at https://github.com/Necla/StemnesScoRe and includes documentation with implementation of a real-data experiment.