Making waves in Breedbase: An integrated spectral data storage and analysis pipeline for plant breeding programs

Abstract

Visible and near-infrared spectroscopy (Vis-NIRS) is a promising tool for increasing phenotyping throughput in plant-breeding programs, but existing analysis software packages are not optimized for a breeding context. Additionally, commercial software options are often outside of budget constraints for some breeding and research programs. To that end, we developed an open-source R package, waves, for the streamlined analysis of spectral data with several cross-validation schemes to assess prediction accuracy. waves is compatible with a wide range of spectrometer models and performs visualization, filtering, aggregation, cross-validation set formation, model training, and prediction functions for the association of Vis-NIR spectra with reference measurements. Furthermore, we have integrated this package into the Breedbase family of open-source databases, expanding the analysis capabilities of this growing digital ecosystem to a number of crop species. Taken together, the standalone and Breedbase versions of waves enhance the accessibility of tools for the analysis of spectral data during the plant breeding process.This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

Low-cost, handheld near-infrared spectroscopy for root dry matter content prediction in cassava

“…After data collection, all scans were filtered according to Mahalanobis distance. Samples with Mahalanobis distances greater than a cutoff set by a χ 2 -distribution with 331 degrees of freedom ( = 0.05) were removed from the analysis (Johnson & Wichern, 2007) using the waves R package version 0.1.1 (Hershberger et al, 2021) in R version 3.5.2 (R Core Team, 2018). In total, eight scans were removed through this procedure.…”

“…Twelve combinations of common preprocessing methods including standard normal variate (Barnes et al, 1989), first and second derivatives, and Savitzky-Golay polynomial smoothing (Savitzky & Golay, 1964) were applied using the R package waves version 0.1.1 (Hershberger et al, 2021). No clear differences in model performance were found between raw and preprocessed data using any of the preprocessing methods for within-trial predictions (Supplemental Figure S2, Supplemental Table S2), thus raw data were used for all subsequent analyses.…”

“…Counts of overlapping clones between trials are listed in Supplemental Table S3. All CV schemes were implemented in the waves R package version 0.1.1 (Hershberger et al, 2021).…”

“…The copyright holder for this preprint (which this version posted April 30, 2021. ; https://doi.org/10.1101/2021.04.30.441802 doi: bioRxiv preprint 10 Prediction Prediction models were developed using three different algorithms in the waves R package version 0.1.1 (Hershberger et al, 2021): partial least squares regression (PLSR) (H. Wold, 1982;S. Wold et al, 1984), random forest (RF) (Breiman, 2001), and support vector machine (SVM) (Drucker et al, 1997).…”

Prospector: A mobile application for portable, high‐throughput near‐infrared spectroscopy phenotyping

“…Prospector provides fast, reliable capture of scans for phenotype prediction with a simplified user interface. Data exported from Prospector follows community data standards and can be directly imported into the R/waves package developed by Hershberger et al (2021) that has been integrated into BreedBase, a popular breeding database (https://breedbase.org/). By combining these available analytical tools with our optimized mobile application, breeders will be able to rapidly adopt NIRS for phenotyping, thereby streamlining complex selection decisions and improving plant breeding productivity.…”