Development of a Long-Read Sequencing Pipeline for Discovering ALS-Associated Genomic Variants Using Adaptive Sampling with Nanopore Technology
Year: 2025
Type: dissertation
Abstract: Most cases of amyotrophic lateral sclerosis (ALS) remain genetically unexplained, highlighting the need for sequencing approaches capable of detecting complex genomic features such as structural variants and repeat expansions. Long-read sequencing used in Oxford Nanopore Technologies (ONT), overcomes the limitations of short-read sequencing and enables targeted enrichment via adaptive sampling. This study aimed to develop a long-read sequencing pipeline for ALS research and to design a custom gene panel suitable for future adaptive sampling applications. A complete ONT-based whole-genome sequencing workflow was implemented, including quality control with NanoPlot, read filtering in Chopper, alignment with Minimap2, and variant-calling with Clair3 (single nucleotide variations), Sniffles2 (structural variations), Spectre (copy number variations) and NanoRepeat (repeat expansions). Annotation was performed in SnpEff, SnpSift, and ClinVar. The pipeline was refined using the control sample GM24149, for high-quality performance. The pipeline was then applied to two ALS patient samples. Patient-1 had suboptimal library preparation, resulting in low coverage (~2.4×) and limited variant detection. Patient-2, with an optimized protocol yielded improved results (~12× coverage), 124,641 single nucleotide variations, and 45,487 structural variants identified. No ClinVar annotation was performed due to moderate depth, but overall variant detection was robust. No pathogenic repeat G4C2-expansion was detected in the C9orf72 locus, which yielded consensus of 7 repeats. A custom ALS panel of 342 genes was designed and validated, with over 150 genes from the ALS Online Database, of which 44 have firm evidence of ALS association and 106 are considered suspected ALS-related; 14 from GWAS studies; and additional loci identified through in-house transcriptomic analysis in ALS mouse models. The BED file covers ~0.6% of the human genome, complying with adaptive sampling indications, supporting barcoding for multiplexed, cost-effective patient sequencing. These results provide a validated framework for future studies aiming to uncover unresolved ALS heritability through targeted long-read sequencing.
Author: Radlowski Nova, Jorge
Language: English
Subfield: Molecular Biology
Domain: Life Sciences
Open Access status: green
OpenAlex ID: W7131337672
Cites:
Cited by:
Related to: