Blackmon Lab · Texas A&M University

AI in
Evolutionary Biology

We use AI tools to accelerate research in evolutionary biology. Our work spans autonomous literature-mining agents, AI-assisted theoretical derivations, and practical curricula for biology students and researchers.

How we use AI in the lab

We use AI as a research tool, not just a productivity shortcut. That means autonomous pipelines with validated outputs, workflows that scale to problems a single lab couldn't otherwise tackle, and curricula that give students practical skills alongside the critical thinking to evaluate AI outputs.

🤖

Autonomous agents

Software agents that search databases, triage papers, extract structured data, and generate new queries based on what they find. They run unattended and pick up where they left off across sessions.

🧠

AI as lead investigator

We are testing whether a Claude-powered agent can act as principal investigator on a complete research project: forming hypotheses, selecting comparative methods, running analyses, and drafting a manuscript. The goal is to understand where AI reasoning works in biology and where it fails.

🎓

AI-literate biologists

Biology students need practical fluency with AI tools alongside the critical thinking to evaluate their outputs. We have built curricula at the undergraduate and graduate levels at Texas A&M to develop both.

📊

Data at scale

Cytogenetics data from the past 80 years are scattered across hundreds of journals in multiple languages. AI makes it feasible to extract, standardize, and analyze this literature at a scale that wouldn't be practical otherwise.

🔬

Validated outputs

Every AI extraction gets a confidence score. Flagged records are routed to domain experts. Validation is built into the pipeline design, not added as an afterthought.

🌐

Open and reproducible

All databases, code, and workflows are published openly so other comparative biology labs can replicate and build on them.

Research projects

Active and recent projects using AI tools in the lab.

📈

RateScape

Active

Penalized likelihood method for estimating branch-specific rate scalars on phylogenies. We used AI tools throughout method development, simulation, and manuscript writing. Planned application to pathogen drug-resistance rate estimation (NHGRI PAR-25-228).

R packages
🗂️

Karyotype Databases

Published

Six interactive karyotype databases covering Coleoptera, Diptera, Amphibia, Mammalia, Drosophila, and Polyneoptera. Over 20,000 records total. All static, downloadable, and machine-readable.

Browse databases
🧬

Population Genetics Simulator

Published

Interactive Wright-Fisher simulator supporting drift, selection, mutation, migration, and bottlenecks. Built entirely through AI-assisted coding as a teaching tool and a test of what a coding agent can produce without manual intervention.

Open simulator
🔣

Sex-Autosome Fusion Theory

Manuscript in prep

Theory paper on sex-autosome fusion fixation probability. We used AI tools for symbolic derivations, analytical cross-checks, and drafting. A useful test case for how AI assistance affects throughput and error rate in theoretical work.

Research page

Teaching

Courses and curricula at Texas A&M giving biology students hands-on experience with AI tools.

🧪

Biology and AI CURE

Spring 2026

Undergraduate research course where each student conducts an original evolutionary biology project using phylogenetic comparative methods and AI-assisted analysis. Spring 2026 cohort.

Meet the cohort
🎓

AI in Biology Concentration

Curriculum

Formal 10-credit-hour concentration in the Texas A&M Biology BS and PhD programs. Built to develop skills that transfer across tools, not just familiarity with whatever is current.

View concentration
📖

AI Tools & Prompting Guides

Open access

Practical guides to AI tools and prompting techniques written for biologists. Covers literature review, data analysis, coding assistance, and writing workflows.

Browse guides

Principles

Some constraints we try to stick to.

01
Validate before trusting

Every AI output is verified before it enters a publication, dataset, or decision. The form of verification — computational check, statistical test, expert review — depends on the task, but the expectation is universal.

02
AI amplifies effort; researchers supply judgment

Search, extraction, reformatting, summarizing, and consistency checking are good uses of AI. Deciding what question is scientifically worth asking, whether a result makes biological sense, and what to conclude — those stay with the researcher.

03
Document how the result was produced

Reproducible science requires knowing not just what AI generated, but what it was asked, with what model, and under what constraints. Prompts, model versions, and pipeline configurations are part of the methods section.

04
Characterize failure modes, not just capabilities

We care as much about documenting where AI reasoning breaks down as demonstrating what it can do. Systematically testing the limits of a tool is as scientifically valuable as deploying it successfully.

05
Make domain knowledge explicit

A general-purpose AI applied to a specialized scientific problem produces generic results. Effective use requires encoding expert knowledge in prompts, constraints, and validation rules — not assuming the model already carries it.

Join the lab

We are recruiting graduate students and postdocs interested in using AI tools in evolutionary biology research. Get in touch if that sounds like you.