Whole-genome sequencing of extrachromosomal circular DNA of cerebrospinal fluid of medulloblastoma

Abstract

BackgroundMedulloblastoma (MB) is a malignant tumor associated with a poor prognosis in part due to a lack of effective detection methods. Extrachromosomal circular DNA (eccDNA) has been associated with multiple tumors. Nonetheless, little is currently known on eccDNA in MB.MethodsGenomic features of eccDNAs were identified in MB tissues and matched cerebrospinal fluid (CSF) and compared with corresponding normal samples using Circle map. The nucleotides on both sides of the eccDNAs’ breakpoint were analyzed to understand the mechanisms of eccDNA formation. Bioinformatics analysis combined with the Gene Expression Omnibus (GEO) database identified features of eccDNA-related genes in MB. Lasso Cox regression model, univariate and multivariate Cox regression analysis, time-dependent ROC, and Kaplan–Meier curve were used to assess the potential diagnostic and prognostic value of the hub genes.ResultsEccDNA was profiled in matched tumor and CSF samples from MB patients, and control, eccDNA-related genes enriched in MB were identified. The distribution of eccDNAs in the genome was closely related to gene density and the mechanism of eccDNA formation was evaluated. EccDNAs in CSF exhibited similar distribution with matched MB tissues but were differentially expressed between tumor and normal. Ten hub genes prominent in both the eccDNA dataset and the GEO database were selected to classify MB patients to either high- or low-risk groups, and a prognostic nomogram was thus established.ConclusionsThis study provides preliminary evidence of the characteristics and formation mechanism of eccDNAs in MB and CSF. Importantly, eccDNA-associated hub genes in CSF could be used as diagnostic and prognostic biomarkers for MB.

eccDNA-pipe: an integrated pipeline for identification, analysis and visualization of extrachromosomal circular DNA from high-throughput sequencing data

“…We collected 42 human muscle or human blood Circle-seq short-read sequencing data [ 38 ], 4 human multiple myeloma cancer cell line data (EJM, JJN3, APR1_rep1, APR1_rep2) and 3 mouse cell line data (5TMG1 multiple myeloma cells, E0771 breast cancer cells, MLOY4 osteocyte-like cells) in Circle-seq long-read sequencing [ 21 ], GBM39 WGS short-read sequencing data [ 8 ] and MB Circulome-seq short-read sequencing data [ 39 ] to test the eccDNA-pipe. For the upstream analysis, we modified parameters in the config file and executed trim_galore to remove adapters from the data.…”

“…For the upstream analysis, we modified parameters in the config file and executed trim_galore to remove adapters from the data. In a similar manner, we adjusted and refined the parameters to execute Circle-Map for circle-seq short-read sequencing data [ 38 ] and MB Circulome-seq short-read sequencing data [ 39 ]. Subsequently, we executed AmpliconArchitecture on GBM39 WGS short-read sequencing data [ 8 ] and executed CReSIL for EJM, JJN3, APR1_rep1, APR1_rep2, 5TMG1, E0771 and MLOY4 circle-seq long-read sequencing data [ 21 ].…”

“…WGS short-read sequencing data of GBM39 was downloaded from GEO: PRJNA506071 [ 8 ]. Circulome-seq short-read sequencing data of medulloblastoma (MB) were downloaded from GEO: GSE205178 [ 39 ].…”

Comparative analysis of methodologies for detecting extrachromosomal circular DNA

“…Lastly, the potential of eccDNA as a diagnostic marker for diseases like advanced chronic kidney disease 25 , medulloblastoma 26 , and colorectal cancer 27 is promising. However, the time-consuming enrichment step, particularly the linear DNA digestion process, may limit the practicality of eccDNA in clinical diagnostics.…”

Long extrachromosomal circular DNA identification by fusing sequence-derived features of physicochemical properties and nucleotide distribution patterns

“…The identification of eccDNA is accomplished using a variety of wet-lab experimental methods, including pulsed-field gel electrophoresis (PFGE) 17 , southern blotting 18 , whole genome sequencing (WGS) 19 , fluorescence in situ hybridization (FISH) 20 , RT-PCR 18 , electron microscopy 21 , and rolling circle amplification (RCA) 22 . However, these methods often require prior knowledge or specific probes capable to bind with eccDNA which can limit their applicability to previously characterized eccDNAs.…”