Recent Publications
Summary measures of predictive power associated with logistic regression models of disease risk
Many people use R2 summary statistics for ordinary linear regression analyses. However, you can’t use the regular types of R2 methods for non-linear regressions, like those that you would use for logistic regression. There are several different R2-analogues out there, but they all do different things and can be useful in different ways. In this paper, we used spore trap data from my PhD thesis to understand how two different metrics work, and what they can tell us about the data. This paper is the culmination of a workshop that Neil and I held at ICPP in 2018; you can find the material for that workshop at https://robchoudhury.github.io/InformationTheory/index.html .
Phytopathology 109: In Press,
2018
Climate change, carbon dioxide, and pest biology, managing the future: coffee as a case study
Climate change is affecting how plants interact with pests and pathogens. Karen and I gave a talk on how changing weather patterns might affect laurel wilt disease, and we were invited to contribute to a summary paper that focused on how CO2 and climate change might impact different things, with a focus on coffee as a case study.
Agronomy 8:152,
2018
Network analysis: a systems framework to address grand challenges in plant pathology
Network analysis can help to address a lot of different problems within plant pathology. In this review, we highlight some uses of networks, and discuss their future.
Annual Review of Phytopathology 56:25.1-25.22,
2018
Multiple origins of downy mildews and mitonuclear discordance within the paraphyletic genus Phytophthora
The phylogenetic relationship between downy mildews and Phytophthoras has always been kind of cryptic. Are they all in the same genus? Closely related genera? Tyler and I deeply wanted these questions answered. He loves Phytophthora, and was entangled in a project to figure out how different forest pathogens were related, and I wanted to know how downy mildews tied in to this important group. We were able to wrangle up some samples and DNA from various downy mildews from all over California, and figured out that 1) the downy mildews are polyphyletic (they arose multiple times, 2) all nested within the ‘genus’ Phytophthora, and 3) that there seems to be differences between the mitochondrial phylogeny and nuclear phylogeny, suggesting there might be some horizontal mitochondrial movement between clades.
PLOS One,
2018
A framework for optimizing phytosanitary thresholds in seed systems
Its often challenging to set a complete, no-holds-barred phytosanitary thresholds. Theyre costly and they can sometimes backfire, resulting in insufficient seeds necessary for a planting season. Setting a non-zero threshold can allow for a reduction in the amount of inoculum imported while still only reducing the number of useable seedlots by a fraction. Here we outline a framework for setting non-zero phytosanitary thresholds.
Phytopathology 107: 1219-1228,
2017
