RGB and hyperspectral phenomics dataset for in vitro transformation and regeneration of Populus trichocarpa
Michael F. Nagle1, Jialin Yuan2, Damanpreet Kaur2, Cathleen Ma1, Ekaterina Peremyslova1, Yuan Jiang3, Greg S. Goralogia1, Anna Magnuson1, Jia Yi Li2, Wellington Muchero4,5,6, Li Fuxin2, and Steven H. Strauss1
1 Department of Forest Ecosystems & Society, Oregon State University, Corvallis, Oregon, 97331, USA
2 School of Electrical Engineering and Computer Science, Oregon State University, Corvallis, Oregon, 97331, USA
3 Statistics Department, Oregon State University, Corvallis, Oregon, 97331, USA
4 Biosciences Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee, 37830, USA
5 Center for Bioenergy Innovation, Oak Ridge National Laboratory, Oak Ridge, Tennessee, 37830, USA
6 Bredesen Center for Interdisciplinary Research, University of Tennessee, Knoxville, Tennessee, 37996, USA
1,209 genotypes of Populus trichocarpa underwent an in vitro transformation and regeneration protocol and were imaged with RGB and hyperspectral cameras (via macoPhor Array, Middleton Spectral Vision). Fluorescent-hyperspectral images enable the visualization of enhanced GFP (eGFP) used as a reporter for successful transformation. This dataset contains 19,961 Petri dish images, spanning three timepoints (after 3 weeks of callus induction, after a subsequent 4 weeks of shoot induction, and again after another 3 weeks of shoot induction). Each genotype is represented by a median of two replicates each of samples with and without an added acetosyringone pre-treatment. The GMOdetector system (https://github.com/naglemi/GMOnotebook) was used to obtain segmentation maps of regenerating tissues (e.g., callus, shoot) from RGB images and cross-reference these with reporter protein signals from fluorescent-hyperspectral images, ultimately producing statistics on regeneration and transformation frequencies of specific tissues. These statistics were then analyzed in a downstream genome-wide association study (GWAS) and tests for genetic interactions among GWAS candidates.