Population-scale long-read DNA sequencing: peering under the hood of the new evolutionary genomics

Abstract

Population-scale long-read DNA sequencing (PLRS) is rapidly reshaping our understanding of genomic variation in humans and non-model species. In this Darwin Review, we first recount the expansion of the PLRS concept and its twin paradigm, the pangenome, over the past 20 years, emphasizing recent results from non-human vertebrates. Using recent PLRS studies in birds as test cases, we probe three aspects of PLRS studies-diploid genome assembly, characterization and annotation of the repeatome, and detection of gene copy number variants-that are being re-shaped by new data types and computational tools. We argue that, in the absence of data from family trios, partially phased haplotypes provide a natural substrate for pangenome analysis, especially when quantifying structural variants directly from pangenome graphs. We identify gaps and discrepancies in the annotation of the repeatome of long-read assemblies, especially for satellite and low-complexity DNA, that are being ameliorated by new computational tools. Finally, we discuss and evaluate the surprising extent of gene copy number variation exposed in recent PLRS studies. The current methodological heterogeneity of pangenome studies may soon coalesce around a few core protocols to the extent allowed by rapidly changing sequencing technologies, allowing greater consistency among studies.