# Causes and consequences of experimental variation in Nicotiana benthamiana transient expression

## TL;DR

This paper investigates and models the variability in a common plant biology technique using Nicotiana benthamiana, offering guidelines to improve reproducibility.

## Contribution

The study quantifies sources of variation in agroinfiltration and provides practical guidelines for minimizing it through experimental design.

## Key findings

- Variability in agroinfiltration experiments is substantial and not consistently reduced by normalization methods.
- A model using power analysis determines the number of plants needed for specific effect sizes in experiments.
- Longitudinal data over three years informs strategies to minimize variability in transient expression studies.

## Abstract

Infiltration of Agrobacterium tumefaciens into Nicotiana benthamiana has become a foundational technique in plant biology, enabling efficient delivery of transgenes in planta with technical ease, robust signal, and relatively high throughput. Despite transient expression’s prevalence in disciplines such as synthetic biology, little work has been done to describe and address the variability inherent in this system, a concern for experiments that rely on highly quantitative readouts. In a comprehensive analysis of N. benthamiana agroinfiltration experiments, we model sources of variability that affect transient expression. Our findings emphasize the need to validate normalization methods under the specific conditions of each study, as distinct normalization schemes do not always reduce variation either within or between experiments. Using a dataset of 1915 plants collected over three years, we develop a model of variation in N. benthamiana transient expression, using power analysis to determine the number of individual plants required for a given effect size. Drawing on our longitudinal data, these findings inform practical guidelines for minimizing variability through strategic experimental design and power analysis, providing a foundation for more robust and reproducible use of N. benthamiana in quantitative plant biology and synthetic biology applications.

Little work has been done to describe and address the variability inherent in the agroinfiltration and genetic engineering of Nicotiana
benthamiana. Here, the authors identify and quantify the sources of virtually all variation and develop recommendations for minimizing variation.