ARS SCINet and AI Center of Excellence Postdoctoral Fellowships Program (FY24)

Overview

The SCINet Program, in collaboration with the ARS Artificial Intelligence Center of Excellence (AI-COE), is calling for proposals for funding to support postdoctoral fellows to be mentored by ARS scientists. The goal of the fellowships program is to develop the next generation of ARS scientists with expertise in conducting and leading individual and collaborative research using computationally intensive approaches.

Each fellow should: (1) be involved in individual and collaborative, multi-unit research that includes substantial computational work and that will leverage SCINet’s computing infrastructure; (2) have training and leadership opportunities; and (3) contribute to the overall success of SCNet or the AI-COE and the SCINet/AI-COE Fellowships Program. In addition, each fellow will have the opportunity to take advantage of training courses that build computational literacy, such as in data science, AI, bioinformatics, and geospatial analyses on SCINet’s high-performance computing clusters (Ceres, Atlas).

All fellows will receive a competitive stipend and travel support. Fellowships are funded for two years. The mentor is expected to provide the fellow’s laptop, office space, and laboratory space (if needed).

Proposal format and submission

All proposals must be submitted using the online submission form. The PI’s RL or supervisor must approve the proposal prior to submission (approval will be indicated on the submission form).

All proposals must use the fellowships proposal template. Please follow the instructions in the template and do not change the section headings or document formatting. The final proposal must be no more than 750 words, including the headings.

Deadline for proposal submission: Close of business on Friday, December 15, 2023.

Eligibility: ARS Category 1, 4, or 6 scientists with RL or supervisor approval. Please note that lead mentors on FY23 awards are not eligible for an FY24 award.

Proposals funded in FY23

Title	Mentor	Co-mentor(s)
Determining the Evolutionary Potential of the Causal Agent of Coffee Leaf Rust Disease From Big Data	Rachel A. Koch Bach
Solving Big Data Challenges in California’s Tree Nut Pests	Raman Bansal
Developing A Cross-Kingdom Gene Editing Tool Kit For USDA-ARS Scientists	Paula Chen	Norman Best, David Kang, Bethany Redel, Adam Rivers, Jacob Washburn
AI Modeling of Grazingland Animal Behavior	Patrick Clark
Networking local measurements to management scales using machine learning	Octavia Crompton	Kyle Knipper
Machine learning reveals RNAi molecular targets from transcriptomic architectures of mycotoxigenic fungi.	Milton Drott	Mitch Elmore, Hye-Seon Kim
Functional genomics and microbes for the Beenome100 pollinator project	Jay Evans	Judy Chen
Food Security Agency in Alaska	Claire Friedrichsen	Katie Pisarello, Dave Archer
Spatiotemporal effects of restoration on soil moisture dynamics in semiarid ecosystems	Alexander J. Hernandez
Using AI to address large, complex datasets in microbiome-based IPM.	David Kang
Using A.I. and Metatranscriptomics to Improve Predictions of Enzyme Function in the Gut Microbiome	Danielle G. Lemay
Identification of fungal proteins and chemicals targeting insect pests	Brian Lovett	Kathryn Bushley
Transforming ARS scientists use of synthetic image creation pipelines for computer vision and AI	Steven Brian Mirsky	Chris Reberg-Horton
A supervised machine learning tool to predict parameters that are correlated with a reduction in antimicrobial resistance and probiotic administration	Adelumola Oladeinde	Zaid Abdo
Engineering agricultural products using an AI/ML approach to study insect enzyme and substrate interactions	Brenda Oppert	Gerard Lazo, Chris Mattison
Machine learning to distinguish pest from non-pest weevils	Lindsey Perkin
Machine Learning Approaches to Improve Functional Predictions of Chemosensory Genes in Stored Product Insects	Erin Scully
Implications of taxon-specific training sets on the accuracy of Google DeepVariant variant calling in non-model (non-human) systems for improved marker-assisted breeding and trait mapping.	Sheina Sim	Scott Geib
ASFV-Swine Reactome: Adapting Deep Learning to Predict Novel Protein Interactions	Edward Spinard
Metagenomic analysis of winegrape soils to develop a unique approach to evaluate soil health outcomes	Kerri Steenwerth	Amisha Poret-Peterson, Cristina Lazcano, Jorge Rodgrigues
Robust explainable AI methods for understanding climate impacts on crop yields	Xuesong Zhang	Glenn E. Moglen, Kaiguang Zhao
Improve Aerial Image-based Forest Fire Detection Using Deep Learning	Huihui Zhang	David Barnard