Kimberley Billingsley from the NIH Center for Alzheimer’s and Related Dementias (CARD) presented at London Calling 2023 on “Population-scale nanopore sequencing to further understand the genetics of Alzheimer’s disease and related dementias.” They described how the center’s mission is to support research for Alzheimer’s and other diseases. The center begins with genetics research. Billingsley noted that Alzheimer’s Disease (AD) is the most common form of dementia followed by Lewy Body Dementia and frontotemporal dementia. Billingsley and colleagues created the CARD Long Read Sequencing Initiative to sequence ~4000 human brain samples and assess the structural variants in AD and related diseases. The added benefits is that they would obtain information on methylation. The team began by establishing a wet-lab protocol for high-throughput Nanopore long-read sequencing followed by implementing a scalable computation pipeline to generate de novo assemblies and storing files. All of their protocols are on Protocols.io! Their N50 is ~30 Kb. The team has developed protocols for brain, frozen blood, and frozen blood pellets. The protocols use the King Fisher Apex for DNA extraction. They also use the PacBio Nanobind reagents. The team sequences with 2-3 library loads per sample. The team has two P48 instruments! Once sequenced, the basecalling happens on Google Cloud for ~$130. The scalable Nanopore sequencing protocol is available as a pre-print and on Github. The team has called it the CARD Terra Workflow. The sequences are assembles after super accurate basecalling. The workflow uses Minimap2. The workflow also allows for haplotype-specific methylation profiling. The CARD Terra Workflow allows the team and others to do high-throughput sequencing of samples. Utilizing samples from clinical trials/cohorts, they have already sequenced a couple hundred patient samples. Moving forward, they will use SQK-LSK 114 and R10.4.1 flow cells. The team has optimized the protocol on the Hamilton liquid handler. They can now prepare 48 samples in four hours! The team is also using Sniffles2 to detect somatic variations. The group is using PacBio too for several samples and brain sites. I enjoyed learning how the team increased throughput and automated the LSK 114 kit.
