On the use of partial area under the ROC curve for comparison of two diagnostic tests

Evaluation of diagnostic performance is typically based on the receiver operating characteristic (ROC) curve and the area under the curve (AUC) as its summary index. The partial area under the curve (pAUC) is an alternative index focusing on the range of practical/clinical relevance. One of the problems preventing more frequent use of the pAUC is the perceived loss of efficiency in cases of noncrossing ROC curves. In this paper, we investigated statistical properties of comparisons of two correlated pAUCs. We demonstrated that outside of the classic model there are practically reasonable ROC types for which comparisons of noncrossing concave curves would be more powerful when based on a part of the curve rather than the entire curve. We argue that this phenomenon stems in part from the exclusion of noninformative parts of the ROC curves that resemble straight‐lines. We conducted extensive simulation studies in families of binormal, straight‐line, and bigamma ROC curves. We demonstrated that comparison of pAUCs is statistically more powerful than comparison of full AUCs when ROC curves are close to a “straight line”. For less flat binormal ROC curves an increase in the integration range often leads to a disproportional increase in pAUCs’ difference, thereby contributing to an increase in statistical power. Thus, efficiency of differences in pAUCs of noncrossing ROC curves depends on the shape of the curves, and for families of ROC curves that are nearly straight‐line shaped, such as bigamma ROC curves, there are multiple practical scenarios in which comparisons of pAUCs are preferable.     

Disentangling the Association between Statins,Cholesterol, and Colorectal Cancer: A Nested Case-Control Study

BackgroundSeveral prior studies have found an association between statin use and reduced risk of colorectal cancer. We hypothesized that these findings may be due to systematic bias and examined the independent association of colorectal cancer risk with statin use, serum cholesterol, and change in cholesterol concentration.ConclusionsAlthough the risk of colorectal cancer was lower in statin users versus nonusers, no difference was observed among those who continued versus discontinued statin therapy, suggesting the potential for indication bias. The association between decreased serum cholesterol and colorectal cancer risk suggests a cholesterol-lowering effect of undiagnosed malignancy. Clinical judgment should be used when considering causes of cholesterol reduction in patients, including those on statin therapy.     

Imaging in five dimensions: time-dependent membrane potentials in individual mitochondria.

Because of its importance in the chemiosmotic theory, mitochondrial membrane potential has been the object of many investigations. Significantly, however, quantitative data on how energy transduction might be regulated or perturbed by the physiological state of the cell has only been gathered via indirect studies on isolated mitochondrial suspensions; quantitative studies on individual mitochondria in situ have not been possible because of their small size, their intrinsic motility, and the absence of appropriate analytical reagents. In this article, we combine techniques for rapid, high resolution, quantitative three-dimensional imaging microscopy and mathematical modeling to determine accurate distributions of a potentiometric fluorescent probe between the cytosol and individual mitochondria inside a living cell. Analysis of this distribution via the Nernst equation permits assignment of potentials to each of the imaged mitochondrial membranes. The mitochondrial membrane potentials are distributed over a narrow range centered at -150 mV within the neurites of differentiated neuroblastoma cells. We find that the membrane potential of a single mitochondrion is generally remarkably stable over times of 40-80 s, but significant fluctuations can occasionally be seen. The motility of individual mitochondria is not directly correlated to membrane potential, but mitochondria do become immobile after prolonged treatment with respiratory inhibitors or uncouplers. Thus, three spatial dimensions, a key physiological parameter, and their changes over time are all quantitated for objects at the resolution limit of light microscopy. The methods described may be readily extended to permit investigations of how mitochondrial function is integrated with other processes in the intact cell.     

Near ultraviolet light inactivation of dihydroxyacid dehydratase in Escherichia coli

The effects of near ultraviolet (NUV) light on a NUV chromophore-containing oxidant-sensitive enzyme, dihydroxyacid dehydratase (DHAD), were measured in seven strains of Escherichia coli. The strains differed in production of the oxidant-defense enzymes, superoxide dismutases (Fe-SOD and Mn-SOD), and catalases HPI and HPII. With the stress of aerobic growth but without NUV exposure, the strains lacking either Fe or Mn SOD or both SODs had 57%, 25%, and 12%, respectively, of the DHAD-specific activity of the parent (K12) strain. Under the same conditions, the catalase strains that were wild type, overproducing, and deficient had comparable DHAD-specific activities. When aerobic cultures were exposed for 30 min to NUV with a fluence of 216 J/m²/s at 310–400 nm, the percentage decreases in DHAD-specific activities were similar (ranging from 75% to 89%) in strains with none, either, or both SODs missing, and in the catalase-overproducing strain. However, the decreases were only 58% and 52% in the strain with catalase missing and in its parent, respectively. The NUV-induced loss of DHAD enzyme activity was not accompanied by any detectable loss of the DHAD protein as measured by polyclonal antibody to DHAD.     

The ADP/ATP carrier is the 32-kilodalton receptor for an NH2-terminally myristylated src peptide but not for pp60src polypeptide.

Membrane binding of pp60src is initiated via its myristylated NH2 terminus. To identify a candidate pp60src docking protein or receptor in the membrane, a radiolabelled peptide corresponding to the pp60src NH2-terminal membrane binding domain was cross-linked to fibroblast membranes and found to specifically label a 32-kDa protein. This protein was purified by appending an affinity tag to the peptide probe so that the cross-linked complex could be isolated via affinity chromatography. Microsequencing indicated that the 32-kDa protein was the mitochondrial ADP/ATP carrier (AAC). This result was further confirmed by the ability of an antibody to the AAC to immunoprecipitate the cross-linked complex, by the ability of certain inhibitors of the AAC to block cross-linking, and by membrane fractionation to show that complex formation occurred essentially exclusively in the mitochondrial fraction. While the AAC bound the myristyl-src peptide in a specific manner both in vitro and in vivo, its localization to the inner membrane of the mitochondrion precludes its being a pp60src binding protein. An analysis of pp60v-src binding in vitro was consistent with this expectation. Thus, use of a myristyl-src peptide revealed an unexpected and previously unidentified binding capacity of the AAC, most likely related to the ability of long-chain fatty acyl coenzyme As to serve as AAC inhibitors. The amphipathic nature of the pp60src NH2 terminus suggests alternative strategies for uncovering pp60src membrane binding species.