Initially described as a triad of immunodeficiency, congenital heart defects and hypoparathyroidism, 22q11.2 deletion syndrome (22q11.2DS) now encompasses a great amount of abnormalities involving different systems. Approximately 85% of patients share a 3 Mb 22q11.2 region of hemizygous deletion in which 46 protein-coding genes are included. However, the hemizygosity of the genes of this region cannot fully explain the clinical phenotype and the phenotypic variability observed among patients. Additional mutations in genes located outside the deleted region, leading to “dual diagnosis”, have been described in 1% of patients. In some cases, the hemizygosity of the 22q11.2 region unmasks autosomal recessive conditions due to additional mutations on the non-deleted allele. Some of the deleted genes play a crucial role in gene expression regulation pathways, involving the whole genome. Typical miRNA expression patterns have been identified in 22q11.2DS, due to an alteration in miRNA biogenesis, affecting the expression of several target genes. Also, a methylation epi-signature in CpG islands differentiating patients from controls has been defined. Herein, we summarize the evidence on the genetic and epigenetic mechanisms implicated in the pathogenesis of the clinical manifestations of 22q11.2 DS. The review of the literature confirms the hypothesis that the 22q11.2DS phenotype results from a network of interactions between deleted protein-coding genes and altered epigenetic regulation.

Parent-of-origin-dependent gene expression of a few hundred human genes is achieved by differential DNA methylation of both parental alleles. This imprinting is required for normal development, and defects in this process lead to human disease. Induced pluripotent stem cells (iPSCs) serve as a valuable tool for in vitro disease modeling. However, a wave of de novo DNA methylation during reprogramming of iPSCs affects DNA methylation, thus limiting their use. The DNA methyltransferase 3B (DNMT3B) gene is highly expressed in human iPSCs; however, whether the hypermethylation of imprinted loci depends on DNMT3B activity has been poorly investigated. To explore the role of DNMT3B in mediating de novo DNA methylation at imprinted DMRs, we utilized iPSCs generated from patients with immunodeficiency, centromeric instability, facial anomalies type I (ICF1) syndrome that harbor biallelic hypomorphic DNMT3B mutations. Using a whole-genome array-based approach, we observed a gain of methylation at several imprinted loci in control iPSCs but not in ICF1 iPSCs compared to their parental fibroblasts. Moreover, in corrected ICF1 iPSCs, which restore DNMT3B enzymatic activity, imprinted DMRs did not acquire control DNA methylation levels, in contrast to the majority of the hypomethylated CpGs in the genome that were rescued in the corrected iPSC clones. Overall, our study indicates that DNMT3B is responsible for de novo methylation of a subset of imprinted DMRs during iPSC reprogramming and suggests that imprinting is unstable during a specific time window of this process, after which the epigenetic state at these regions becomes resistant to perturbation.

CRC is an adult-onset carcinoma representing the third most common cancer and the second leading cause of cancer-related deaths in the world. EO-CRC (<45 years of age) accounts for 5% of the CRC cases and is associated with cancer-predisposing genetic factors in half of them. Here, we describe the case of a woman affected by BWSp who developed EO-CRC at age 27. To look for a possible molecular link between BWSp and EO-CRC, we analysed her whole-genome genetic and epigenetic profiles in blood, and peri-neoplastic and neoplastic colon tissues. The results revealed a general instability of the tumor genome, including copy number and methylation changes affecting genes of the WNT signaling pathway, CRC biomarkers and imprinted loci. At the germline level, two missense mutations predicted to be likely pathogenic were found in compound heterozygosity affecting the Cystic Fibrosis (CF) gene CFTR that has been recently classified as a tumor suppressor gene, whose dysregulation represents a severe risk factor for developing CRC. We also detected constitutional loss of methylation of the KCNQ1OT1:TSS-DMR that leads to bi-allelic expression of the lncRNA KCNQ1OT1 and BWSp. Our results support the hypothesis that the inherited CFTR mutations, together with constitutional loss of methylation of the KCNQ1OT1:TSS-DMR, initiate the tumorigenesis process. Further somatic genetic and epigenetic changes enhancing the activation of the WNT/beta-catenin pathway likely contributed to increase the growth advantage of cancer cells. Although this study does not provide any conclusive cause–effect relationship between BWSp and CRC, it is tempting to speculate that the imprinting defect of BWSp might accelerate tumorigenesis in adult cancer in the presence of predisposing genetic variants.

Silver–Russell syndrome is an imprinting disorder characterised by pre- and post-natal growth retardation and several heterogeneous molecular defects affecting different human genomic loci. In the majority of cases, the molecular defect is the loss of methylation (LOM) of the H19/IGF2 differentially methylated region (DMR, also known as IC1) at the telomeric domain of the 11p15.5 imprinted genes cluster, which causes the altered expression of the growth controlling genes, IGF2 and H19. Very rarely, the LOM also affects the KCNQ1OT1 DMR (also known as IC2) at the centromeric domain, resulting in an SRS phenotype by an unknown mechanism. In this study, we report on two cases with SRS features and a LOM of either IC1 and IC2. In one case, this rare and complex epimutation was secondary to a de novo mosaic in cis maternal duplication, involving the entire telomeric 11p15.5 domain and part of the centromeric domain but lacking CDKN1C. In the second case, neither the no 11p15.5 copy number variant nor the maternal-effect subcortical maternal complex (SCMC) variant were found to be associated with the epimutation, suggesting that it arose as a primary event. Our findings further add to the complexity of the molecular genetics of SRS and indicate how the LOM in both 11p15.5 DMRs may result from different molecular mechanisms.

Motivation: Inferring gene networks from single-cell RNA sequencing data is challenging due to high sparsity, dimensionality, and technical noise. Current pipelines lack the multi-dataset integration and comprehensive post-processing analysis. Results: scGraphVerse is an R package that integrates multiple algorithms (GENIE3, GRNBoost2, ZILGM, PCzinb, and JRF) with extensive evaluation and visualization tools. Its modular workflow supports early, late, and joint integration strategies for multi-dataset analysis, providing standardized input/output interfaces and biological interpretation tools, including community detection, pathway enrichment, and literature mining. Benchmarking on simulated data showed model-based methods (PCzinb and ZILGM) perform well with limited sample sizes, while JRF performs best as the network size and dataset numbers increase. A PBMC case study demonstrates JRF’s ability to identify literature-supported regulatory communities across donors. Availability and implementation: The package is available in Bioconductor 3.22 at https://bioconductor.org/packages/release/bioc/html/scGraphVerse.html. Code and examples: https://github.com/ngsFC/scGV_analysis.

Background/Objectives: Acute lymphoblastic leukaemia (ALL) is a biologically heterogeneous disease in which transcriptional dysregulation contributes to disease onset and progression. Despite survival rates exceeding 90% in high-income countries, relapsed and high-risk cases remain a major clinical challenge, highlighting the need for improved molecular stratification, namely the classification of patients based on genetic and transcriptomic features associated with prognosis, therapeutic response, and disease biology, as well as for the identification of novel therapeutic targets. Methods: We performed an integrative cross-platform analysis to investigate the expression and potential relevance of three candidate genes: PXDN, TCF4, and TSPAN7 in ALL. Gene expression was interrogated across the MILE microarray cohort and the St. Jude Cloud PeCan paediatric RNA-sequencing dataset. Results: Differential expression analyses consistently showed significant upregulation of TCF4 and PXDN in B-cell ALL (B-ALL) across both platforms (adjusted p < 0.001), while TSPAN7 displayed higher expression in T-cell ALL (T-ALL) and variable upregulation in B-ALL. These findings were supported by preliminary validation using quantitative PCR in paediatric B-ALL samples. To explore potential functional associations, we performed gene regulatory network inference using scGraphVerse, identifying differentially expressed genes putatively linked to PXDN, TCF4, and TSPAN7. Structural modelling using AlphaFold suggested candidate protein–protein interaction interfaces for a subset of these genes, although these predictions require experimental validation. Functional enrichment analysis indicated an over-representation of developmental pathways associated with PXDN- and TCF4-related networks, whereas TSPAN7-associated genes were enriched in processes linked to neuronal lineage development. Conclusions: Collectively, our results identify, for the first time, PXDN, TCF4 and TSPAN7 as differentially expressed genes in ALL and highlight the usefulness of integrative transcriptomic analyses across independent datasets. While limited by small-scale experimental validation and reliance on computational predictions, this study provides a framework for prioritising candidate genes and generates testable hypotheses regarding their potential involvement in leukaemia-associated molecular pathways.