David Haussler and colleagues had been working on how to align a new read library against thousands of human genomes. The easy solution of performing the same alignment thousand times does not scale well. Benedict Paten, Adam Novak and David Haussler posted a new paper in arxiv to address this question. This appears like an interesting paper, but we have not read it carefully to understand the innovation beyond what Juni Siren (corrected) and others proposed before.
To support comparative genomics, population genetics, and medical genetics, we propose that a reference genome should come with a scheme for mapping each base in any DNA string to a position in that reference genome. We refer to a collection of one or more reference genomes and a scheme for mapping to their positions as a reference structure. Here we describe the desirable properties of reference structures and give examples. To account for natural genetic variation, we consider the more general case in which a reference genome is represented by a graph rather than a set of phased chromosomes; the latter is treated as a special case.
Readers may also find the following work from Haussler group relevant -
Speaking of Juni Siren’s algorithm, following twitter discussions are relevant