Before COA

I began my career in environmental science, operating and repairing analytical instruments that measure air- and water-borne pollutants. As I accumulated larger data sets, I found that manual data analysis was no longer practical, which led me to teach myself scripting languages that automate data collection, analysis and reporting. This experience led me into full time programming and I earned a programming certificate, completed some graduate level Computer Science courses, and became a programmer in the Environmental Programs Group at the North Carolina Supercomputing Center. I then earned a Ph.D in Environmental Science and Engineering with a focus on computational biology and genetic influences in environmental toxicology.

Prior to joining the COA faculty I was a Bioinformatics Analyst at The Jackson Laboratory, where I developed computational and statistical methods to identify genes underlying mechanisms of human disease.

Scholarly and Creative Interests

I’m interested in using computational and statistical methods to answer questions that we have about the world. I’m also interested in communicating these answers in a clear and comprehensible manner. These interests fall under the broad heading of Data Science, which is an amalgamation of statistics, machine learning and data visualization. My work focuses on ways to make the analysis methods and results, reliable, reproducible and transparent.

I’m also interested in the broader effect of computers on the human ecosystem. Computer technology has infiltrated every aspect of our lives. Algorithms decide how long we should stay in prison, what medical treatment we should receive, what advertisements we see and our eligibility for loans and jobs. Are these algorithms truly unbiased or do they mask the subjectivity of the input data? What effect does this data have on our lives? How can we evaluate the balance between benefits and harm caused by third parties who use our personal data?

More Information about my Courses

I am developing a curriculum with courses in Data Science and computer programming that will use R, python and javascript. In future years, I plan to develop coursework in bioinformatics, robotics and micro-controllers and web programming.

And yes, the Bicycle course is coming in Spring 2020!

I use Twitter for professional, computer-related posts that I think are worth sharing. @DanielGatti11

More About Me

I once had two huskies, Storm and Sky. They passed from natural causes in the Fall of 2018, but are still in my heart every day.

Storm & Sky on Mt. Cadillac


Genetic background influences susceptibility to chemotherapy-induced hematotoxicity.
Gatti DM, Weber SN, Goodwin NC, Lammert F, Churchill GA. Pharmacogenomics J. 2017 Jun 13. PMID: 28607509

Diversity Outbred Mice Identify Population-Based Exposure Thresholds and Genetic Factors that Influence Benzene-Induced Genotoxicity. French JE, Gatti DM, Morgan DL, Kissling GE, Shockley KR, Knudsen GA, Shepard KG, Price HC, King D, Witt KL, Pedersen LC, Munger SC, Svenson KL, Churchill GA. Environ Health Perspect. 2015 Mar;123(3):237-45. PMID: 25376053

Quantitative trait locus mapping methods for diversity outbred mice. Gatti DM, Svenson KL, Shabalin A, Wu LY, Valdar W, Simecek P, Goodwin N, Cheng R, Pomp D, Palmer A, Chesler EJ, Broman KW, Churchill GA. G3 (Bethesda). 2014 Sep 18;4(9):1623-33. PMID: 25237114

Heading down the wrong pathway: on the influence of correlation within gene sets. Gatti DM, Barry WT, Nobel AB, Rusyn I, Wright FA. BMC Genomics. 2010 Oct 18;11:574. PMID: 20955544

SAFEGUI: resampling-based tests of categorical significance in gene expression data made easy. Gatti DM, Sypa M, Rusyn I, Wright FA, Barry WT. Bioinformatics. 2009 Feb 15;25(4):541-2. PMID: 19098030

FastMap: fast eQTL mapping in homozygous populations. Gatti DM, Shabalin AA, Lam TC, Wright FA, Rusyn I, Nobel AB. Bioinformatics. 2009 Feb 15;25(4):482-9. PMID: 19091771