Frequently asked questions
The basics of what we’re building and why.
What is pZero?
pZero is a global scientific effort to turn insect behavior into training data. Our models learn which kinds of compounds are most likely to change an insect’s decision to land on crops.
What is the key question pZero is attempting to answer?
How do you change an insect’s decision to land on crops?
Who can join pZero?
pZero is being designed for university labs, research institutes, and qualified independent researchers studying insect behavior and crop protection. Participation requirements are still being defined.
How do I apply?
Explore the request access form here. This local preview lets you review the proposed application; it does not submit an application or grant access.
What will pZero provide?
The proposed lab program includes shared protocols, comparable behavioral measurements, and a common research workflow. Compound supply, assay materials, funding, and support arrangements remain to be defined.
What problem are you solving?
We want to understand which compounds can change an insect’s decision to land on crops, and whether those behavioral effects can translate into useful crop protection. That requires evidence from controlled assays and field evaluation.
Who are some of your partners?
pZero’s partner network is still being defined. The institutions named on the M Zero reference site are associated with that program; their participation in pZero has not been established.
Why is this effort even more relevant now?
Machine learning creates a reason to generate experimental evidence in a more comparable form. pZero aims to connect recorded insect behavior, compound information, and crop context so researchers can test whether models improve the choice of the next experiment.
What is the contact/non-contact assay?
A contact assay permits physical contact with a treated surface; a non-contact assay separates the insect from it. See the proposed crop-assay framework and clearly attributed reference material here.
Why begin with one species if crops attract many insects?
A defined species and crop system can make initial experiments easier to compare. pZero’s first species and crops remain to be selected, and findings would need separate validation before being generalized to other systems.
Can I use pZero data in my own research?
pZero’s data-release and licensing terms are still being defined. M Zero uses CC BY 4.0 for shared data; that reference does not establish a license for future pZero datasets.
What paper topics could a participating lab pursue?
- Inter-lab reproducibility of crop-landing assays across sites and insect populations.
- Comparing behavioral responses on treated crops and matched controls.
- Benchmarking predicted changes in landing against prospective assay outcomes.
- Structure–activity relationships in compounds evaluated through a shared assay framework.
- Studies of crop health and beneficial-insect responses alongside behavioral efficacy.
How does the predictive model work?
The proposed model would learn from compound information, measured insect behavior, and assay context. Reviewed results would update its predictions and uncertainty, informing which compound, dose, formulation, or controlled assay adjustment to test next. Its usefulness would be evaluated prospectively: does it help select better experiments and find effects that hold up under new conditions?
How will our data be used?
The intended use is to train and evaluate models that connect compounds with changes in insect behavior. Comparable results, including ineffective compounds and uncertain measurements, would help assess predictions and choose follow-up experiments. Data access, contributor attribution, review, and release terms need to be agreed before the program accepts research data.
How is data quality controlled across labs?
The proposed approach uses a defined protocol for each species and crop system, matched controls, consistent recording conditions, and traceable metadata. Validation across laboratories would test whether results are comparable.
Candidate checks for the initial validation plan include:
- Agreement between replicate assays
- Re-test confirmation of candidate effects
- Variation between laboratories
- Control performance and measurement uncertainty