Extracting abundance information from DNA-based data

The accurate extraction of species-abundance information from DNA-based
data (metabarcoding, metagenomics) could contribute usefully to the
reconstruction of diets and quantitative foodwebs, the inference of
species interactions, the modelling of population dynamics and species
distributions, the biomonitoring of environmental state and change, and
the inference of false positives and negatives. However, capture bias,
capture noise, species pipeline biases, and pipeline noise all combine to
inject error into DNA-based datasets. This review focuses on methods for
correcting the latter two error sources, as the first two are addressed
extensively in the ecological survey literature. To extract abundance
information from DNA-based data, it is useful to distinguish two concepts.
(1) Across-species quantification describes relative species
abundances within a single sample. (2) In
contrast, within-species quantification describes how the
abundance of each individual species varies across samples, where the
samples could be a time series, an environmental gradient, or different
experimental treatments. In the first part of this paper, we review
methods to remove species pipeline biases and pipeline noise. In
the second part, we provide a detailed protocol and demonstrate
experimentally how to use a ‘DNA spike-in’ (an internal standard) to
remove pipeline noise and recover within-species abundance information.