All partnering requests
Synthesizable‑First Generative AI for Crop‑Protection Molecules
  • Background
  • What we're looking for
  • What we can offer you
  • Who we are
  • Q&A
Have questions about this request?
Get them answered by the team at Corteva.
Background

Discovery is slowed by late recognition that proposed molecules are hard to make or underperform at the target. Rule‑based filters and classic QSAR under‑explore chemical space and ignore synthesizability early. Modern generative models—conditioned by structure, ligand data, and rich molecular representations—can propose novel, makeable candidates while predictive models triage by activity and basic developability (including ADMET: absorption, distribution, metabolism, excretion, toxicity). Adding target/structure enablement, calibrated uncertainty, and FTO‑aware novelty triage improves decision‑quality and reduces wasted synthesis.

What we're looking for

We are looking for generative and predictive methods that propose synthesizable small molecules for agricultural targets and robustly prioritize them in silico.

Solutions of interest include:
  • Calibrated uncertainty & domain‑of‑applicability for all predictions
  • Fragment‑based & scaffold‑constrained generation (fragment/RECAP spaces)
  • Generative models with integrated synthesizability constraints
  • Graph & 3D molecular representation learning (equivariant GNNs)
  • Ligand‑based bioactivity–guided generation (QSAR + generative)
  • Multi‑objective molecular optimization (binding, SA, diversity)
  • Predictive ADMET & physicochemical modeling (solubility, logP, pKa)
  • Structure‑based binding affinity prediction (docking + ML rescoring)
  • Structure‑based generative modeling (diffusion/VAEs conditioned on targets)
  • Target/structure enablement for ag targets (homology/cryo‑EM/X‑ray)
Our must-have requirements are:
  • Benchmarking or clear differentiator.
  • Experience with small molecules and protein inhibition.
Our nice-to-have's are:
  • Estimate of time for running through their process to identify small molecule prohibitors (i.e. months, years). Number of iterative cycles to be successful. The ideal timeline would be protein inhibition within the nanomolar range of 6 months.
What's out of scope:
  • Focused on protein inhibitors
  • Overly focused in one area of biology (i.e Oncology or RNAI). Flexibility is needed.
Acceptable technology readiness levels (TRL):
Levels 3-9
What we can offer you
Eligible partnership models:
Fee-for-serviceEquity investmentGift/award
Benefits:
Gift/award
Depending on the proposal quality open to investment or fee for service partnership.
Who we are

At Corteva we use the science of the lab and the land to maximize productivity and sustainability of the world's farmland.

We accomplish this through collaborating with thought leaders and innovators around the world to access and develop the most innovative technologies and rapidly deploying those through organizations that are good stewards of those technologies.

Join us and other thought leaders from around the world to stimulate the development of groundbreaking and sustainable solutions.

 Learn more
Reviewers
Q&A with Corteva

The Q&A is now closed.

Sign up to access the full partnering request.
View the details of this request and connect directly with corporate R&D teams at Fortune 500 companies.
Share partnering request
Complete
Refine recommendations
Is this request relevant to you?
Eligible partners
ConsultantsStartupsSuppliersService providers (CROs/CMOs)
Seeking partners focused on
Agricultural ScienceArtificial IntelligenceBiochemistryComputational ChemistryMachine LearningMedicinal ChemistryPharmaceutical ChemistryPlant PathologyToxicology

Halo home
Partner smarter. Move faster.
Get new partnering requests
delivered to your inbox.