Temporal and spatial variation of nitrogen transformations in nitrogen-saturated soils of a central Appalachian hardwood forest

We studied temporal and spatial patterns of soil nitrogen (N) dynamics from 1993 to 1995 in three watersheds of Fernow Experimental Forest, W.V.: WS7 (24-year-old, untreated); WS4 (mature, untreated); and WS3 (24-year-old, treated with (NH4)2SO since 1989 at the rate of 35 kg Nha–1year–1). Net nitrification was 141, 114, and115 kg Nha–1year–1, for WS3, WS4, and WS7, respectively, essentially 100% of net N mineralization for all watersheds. Temporal (seasonal) patterns of nitrification were significantly related to soil moisture and ambient temperaturein untreated watersheds only. Spatial patterns of soil water NO3–of WS4 suggest that microenvironmental variabilitylimits rates of N processing in some areas of this N-saturated watershed, in part by ericaceous species in the herbaceous layer. Spatial patterns of soil water NO3–in treated WS3 suggest that later stages of N saturation may result inhigher concentrations with less spatial variability. Spatial variability in soil N variables was lower in treated WS3 versus untreated watersheds. Nitrogen additions have altered the response of N-processing microbes to environmental factors, becoming less sensitive to seasonal changes in soil moisture and temperature. Biotic processes responsible forregulating N dynamics may be compromised in N-saturated forest ecosystems.

Classification and selection of biomarkers in genomic data using LASSO

High-throughput gene expression technologies such as microarrays have been utilized in a variety of scientific applications. Most of the work has been on assessing univariate associations between gene expression with clinical outcome (variable selection) or on developing classification procedures with gene expression data (supervised learning). We consider a hybrid variable selection/classification approach that is based on linear combinations of the gene expression profiles that maximize an accuracy measure summarized using the receiver operating characteristic curve. Under a specific probability model, this leads to consideration of linear discriminant functions. We incorporate an automated variable selection approach using LASSO. An equivalence between LASSO estimation with support vector machines allows for model fitting using standard software. We apply the proposed method to simulated data as well as data from a recently published prostate cancer study.