Mathilde Filser from the Curie Institute in France spoke about “Simultaneous methylation and fusion detection for pediatric tumor characterization” at London Calling 2024. They explained that neuro-oncology includes many entities and DNA methylation has an “undeniable impact.” The gold standard is the Illumina EPIC array. The lab developed a workflow for frozen brain tumor samples. The protocol used 2 ug genomic DNA and the Ligation Sequencing Kit (LSK) 114. G-tubes were used to fragment the DNA. Half of the library was injected and sequenced for 14 hours. Then, the flow cell was washed and reloaded to sequence for another 24 hours. The team used the Mk1B device. Adaptive sampling was used to enrich specific regions of interest. The bed file was designed to create a list of fusion genes to sequence. Filser and the team developed a pipeline called NanoCliD that used Dorado for base-calling and demultiplexing. Mapping is performed with Minimap2. SNV calling, variants, and CNV are determined with various programs. Methylation profiling, fusion detection, and whole genome copy number information are obtained. This approach has been applied to ten frozen brain tumors. Filser shared examples, including a one-year-old with a right hemispheric tumor. A fusion was detected, and the tumor was classified. In a second case, an eleven-year-old with a tumor was studied. The tumor was characterized, highlighting a fusion. In the last example, Filser spoke about a sarcoma found in a six-year-old patient. A fusion was detected that allowed classification. Filser summarized that the approach leverages adaptive sampling to provide critical information for tumor characterization with a rapid turnaround of ~11 days. I liked their library prep with 2 ug of DNA (double the typical amount) and wash-reload approach.
