Data describing pollen identified from honey samples originating from the UKCEH National Honey Monitoring Scheme for 2019

All honey samples were submitted following an application though the National Honey Monitoring Scheme online portal which required minimum meta-data on site location and sample date which was verified. This included in some cases additional data on hive health metrics. DNA meta-barcoding to rapidly process samples to identify plant species. Quality assurance of the DNA metabarcoding was delivered through operational deployment of sophisticated protocols for barcoding and interpreting large volumes of honey samples. To do this we developed the HONEYPI pipeline implemented in python 2.7 and is open access (https://github.com/hsgweon/honeypi). The HONEYPI pipeline is divided into several parts as follows: 1) the raw amplicon sequences are quality filtered and adapters removed; 2) DADA2 pipeline is subsequently used to generate an Amplicon Sequence Variant (ASV) abundance table containing chimera-removed, high-quality error-corrected sequences. 3). For each ASV, conserved regions flanking ITS2 are removed; and (4) resulting sequences taxonomically classified using the naive Bayesian classifier against in-house ITS2 database. Since HONEYPI uses ASVs rather than clusters of sequences for classification, it allows combining of ASV tables, i.e. data from two or more separate sequencing runs can be merged without re-clustering sequences.

A full open access methodological paper describing this approach is given in Oliver et al (2021) MethodsX, 8, 101303 (https://doi.org/10.1016/j.mex.2021.101303)