Comparison of small n statistical tests of differential expression applied to microarrays

Background  

DNA microarrays provide data for genome wide patterns of expression between observation classes. Microarray studies often have small samples sizes, however, due to cost constraints or specimen availability. This can lead to poor random error estimates and inaccurate statistical tests of differential expression. We compare the performance of the standard t-test, fold change, and four small n statistical test methods designed to circumvent these problems. We report results of various normalization methods for empirical microarray data and of various random error models for simulated data.     

Fish condition in introduced tilapias of Ugandan crater lakes in relation to deforestation and fishing pressure

This study identifies environmental predictors of the condition of two introduced tilapia species (Oreochromis leucostictus and Tilapia zillii) that are known to have divergent trophic niches (planktivore and herbivore, respectively) in 17 crater lakes in western Uganda. We asked whether fish condition differs among lakes characterized by differences in fishing pressure and catchment deforestation; and we related relative condition factor to gradients of environmental variation across lakes. Lakes characterized by severe catchment deforestation tended to be lakes with high fishing pressure, so it was difficult to explore independent and interactive effects. However, mean relative condition factor was higher in populations with high fishing pressure compared to populations with low fishing pressure for both O. leucostictus and T. zillii. The condition of O. leucostictus populations was higher in lakes with severely deforested catchments; but mean relative condition factor of T. zillii did not differ between deforestation categories. Principal components analysis (PCA) was used to describe the major environmental gradients of variation among the lakes; and PCA factor scores were regressed against relative fish condition. The association between fish condition and environmental gradients was stronger for O. leucostictus than for T. zillii. For O. leucostictus, fish condition was related to PC1 (43% of the variance) and factors that loaded most heavily included Chl-a, water transparency, lake area and depth, suggesting higher condition in lakes characterized by higher primary productivity and smaller size. For T. zillii, PC3 (11%) was the only axis related to fish condition; and factors that loaded most heavily included lake area (positive), and conductivity and total nitrogen (negative). Some of the larger lakes are characterized by higher availability of macrophytes that may positively affect the food base for T. zillii.     

Variable growth and longevity of yellow bullhead (Ameiurus natalis) in the Everglades of south Florida,USA

Yellow bullhead (Ictaluridae: Ameiurus natalis) is the most abundant ictalurid catfish in the Everglades of southern Florida, USA, and, as both prey and predator, is one of many essential components in the ecological‐simulation models used in assessing restoration success in the Everglades. Little is known of its biology and life history in this southernmost portion of its native range; the present study provides the first estimates of age and growth from the Everglades. In total, 144 yellow bullheads of 97–312 mm total length (TL) were collected from canals and marshes of the Everglades between April 2000 and January 2001, and from October 2003 to February 2005. Fish were aged using cross‐sections of pectoral spines and ranged from 1–12 years, with the maximum age almost twice that of any yellow bullhead previously reported. Yellow bullheads from south Florida grew relatively rapidly during their first 3 years, but after age 5 growth slowed and fish approached an asymptote of ～214 mm TL. Compared to other populations in the United States, yellow bullhead in the Everglades grew relatively slowly, were smaller at age overall, but survived to older ages.