Engineering Papers⌕ Search

Engineering topics

Krofcheck, Daniel J.

Publications and source records attributed to Krofcheck, Daniel J..

Data Science and Machine Learning for Genome Security

This report describes research conducted to use data science and machine learning methods to distinguish targeted genome editing versus natural mutation and sequencer machine noise. Genome editing capabilities have been around for more than 20 years, and the efficiencies of these techniques has improved dramatically in the last 5+ years, notably with the rise of CRISPR-Cas technology. Whether or not a specific genome has been the target of an edit is concern for U.S. national security. The research detailed in this report provides first steps to address this concern. A large amount of data is necessary in our research, thus we invested considerable time collecting and processing it. We use an ensemble of decision tree and deep neural network machine learning methods as well as anomaly detection to detect genome edits given either whole exome or genome DNA reads. The edit detection results we obtained with our algorithms tested against samples held out during training of our methods are significantly better than random guessing, achieving high F1 and recall scores as well as with precision overall.

59 BASIC BIOLOGICAL SCIENCES↗

Tree Mortality Decreases Water Availability and Ecosystem Resilience to Drought in Piñon-Juniper Woodlands in the Southwestern U.S.: Tree Mortality in Semiarid Biomes

Climate-driven tree mortality has increased globally in response to warmer temperature and more severe drought. To examine how tree mortality in semi-arid biomes impacts surface water balance, we experimentally manipulated a piñon-juniper (PJ) woodland by girdling all adult piñon trees in a 4 ha area, decreasing piñon basal area by ~65%. Over 3.5 years (2009-2013), we compared water flux measurements from this girdled site with those from a nearby intact PJ woodland. Before and after girdling, the ratio of evapotranspiration (ET) to incoming precipitation was similar between the two sites. Girdling altered the partitioning of ET such that the contribution of canopy transpiration to ET decreased 9-14% over the study period, relative to the intact control, while non-canopy ET increased. We attributed the elevated non-canopy ET in the girdled site each year to winter increases in sublimation, and summer increases in both soil evaporation and below-canopy transpiration. Although we expected that mortality of a canopy dominant would increase the availability of water and other resources to surviving vegetation, we observed a decrease in both soil volumetric water content and sap flow rates in the remaining trees at the girdled site, relative to the control. This post-girdling decrease in the performance of the remaining trees occurred during the severe 2011-2012 drought, suggesting that piñon mortality may trigger feedback mechanisms that leave PJ woodlands drier relative to undisturbed sites, and potentially more vulnerable to drought.

Morillas, L.↗