Applications of de Bruijn graphs in microbiome research

Abstract

High‐throughput sequencing has become an increasingly central component of microbiome research. The development of de Bruijn graph‐based methods for assembling high‐throughput sequencing data has been an important part of the broader adoption of sequencing as part of biological studies. Recent advances in the construction and representation of de Bruijn graphs have led to new approaches that utilize the de Bruijn graph data structure to aid in different biological analyses. One type of application of these methods has been in alternative approaches to the assembly of sequencing data like gene‐targeted assembly, where only gene sequences are assembled out of larger metagenomes, and differential assembly, where sequences that are differentially present between two samples are assembled. de Bruijn graphs have also been applied for comparative genomics where they can be used to represent large sets of multiple genomes or metagenomes where structural features in the graphs can be used to identify variants, indels, and homologous regions in sequences. These de Bruijn graph‐based representations of sequencing data have even begun to be applied to whole sequencing databases for large‐scale searches and experiment discovery. de Bruijn graphs have played a central role in how high‐throughput sequencing data is worked with, and the rapid development of new tools that rely on these data structures suggests that they will continue to play an important role in biology in the future.

iMeta: Integrated meta‐omics for biology and environments

“…Michel‐Mata et al [31] introduced a deep learning algorithm to predict microbiome composition. Reviews are also essential for peers to learn and begin their studies readily. Dufault‐Thompson et al [32] provided a comprehensive and insightful review of de Bruijn graphs' progress and future applications in microbiome studies. Wang et al [33] reviewed plant genetics‐mediated remodeling of microbiota and techniques in response to stress, and Yang et al [34] reviewed the advantages of the targeted approach in integrating metagenome data for accurate protein structure prediction.…”

Extremely-fast construction and querying of compacted and colored de Bruijn graphs with GGCAT

“…One can think of a colored de Bruijn graph as a compressed representation of the k -mers in a collection of datasets, but which retains enough information (i.e., the color of every k -mer) in order to identify each dataset in this combined graph. Applications include [34, 3, 19, 33, 10], just to name a few.…”

“…One can think of a colored de Bruijn graph as a compressed representation of the k -mers in a collection of datasets, but which retains enough information (i.e., the color of every k -mer) in order to identify each dataset in this combined graph. Later applications include pangenomics [37], RNA-seq quantification [5], bacterial genome querying [22], alignment and reference-free phylogenomics [37], microbiome research [13], just to name a few.…”

Extremely fast construction and querying of compacted and colored de Bruijn graphs with GGCAT

“…One can think of a colored de Bruijn graph as a compressed representation of the k -mers in a collection of data sets, but it retains enough information (i.e., the color of every k -mer) in order to identify each data set in this combined graph. Later applications include pangenomics ( Zekic et al 2018 ), RNA-seq quantification ( Bray et al 2016 ), bacterial genome querying ( Luhmann et al 2021 ), alignment and reference-free phylogenomics ( Zekic et al 2018 ), and microbiome research ( Dufault-Thompson and Jiang 2022 ), just to name a few.…”