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 done on assessing univariate associations between gene expression profiles 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 the 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.     

Functional significance of a natural allelic variant of human carbonyl reductase 3 (CBR3).

Human carbonyl reductase (CBR) activity accounts for a significant fraction of the metabolism of endogenous and xenobiotic carbonyl compounds. It is possible that genetic polymorphisms in CBR1 and CBR3 are key for the wide interindividual variability in the disposition of CBR drug substrates. We pinpointed a single nucleotide polymorphism in CBR3 (CBR3 V244M) that encodes for a V244 to M244 change. Blacks showed a higher frequency of the M244 allele (q = 0.51, n = 49) than did whites (q = 0.31, n = 70; p = 0.003). In addition, DNA variation panels from 10 ethnic groups presented a wide range of CBR3 V244M genotype distributions. Kinetic experiments with the recombinant CBR3 protein variants and menadione revealed that CBR3 M244 has significantly higher V(max) than does CBR3 V244 (V(max) CBR3 M244 = 40.6 +/- 1.3 micromol/min x mg versus V(max) CBR3 V244 = 19.6 +/- 2.0 micromol/min x mg, p = 0.002). In contrast, both isoforms presented similar K(m) values (K(m) CBR3 M244 = 22.9 +/- 2.9 microM versus K(m) CBR3 V244 = 24.6 +/- 3.2 microM, p = 0.43). Assays with NADP(H) demonstrated a higher V(maxNADP(H)) (1.6-fold) and increased catalytic efficiency (V(maxNADP(H))/K(mNADP(H))) for CBR3 M244 compared with CBR3 V244 (p = 0.013). Comparative three-dimensional analyses based on the structure of the homologous porcine carbonyl reductase suggested that the V244M substitution is positioned in a region critical for interactions with the NADP(H) cofactor. These studies demonstrate that the common CBR3 V244M polymorphism encodes for CBR3 isoforms with distinctive enzymatic properties.     

Alpha-methylacyl-CoA racemase protein expression is associated with the degree of differentiation in breast cancer using quantitative image analysis.

Alpha-methylacyl-CoA racemase (AMACR) is an enzyme involved in the metabolism of fatty acids and is an important tissue biomarker in the prostate to distinguish normal glands from prostate cancer. Here, for the first time, we evaluated the expression of AMACR protein in normal breast, ductal carcinoma in situ, and invasive carcinomas. By immunofluorescence and immunohistochemistry, AMACR was seen in cytoplasmic granules consistent with a mitochondrial and peroxisomal localization. AMACR expression was determined by immunohistochemistry on 160 invasive carcinomas with long follow-up, using a high-density tissue microarray, and evaluated by two methods: standard pathology review and quantitative image analysis. AMACR was overexpressed in 42 of 160 (26%) invasive carcinomas, and it was associated with a decrease in tumor differentiation, a feature of aggressive breast cancer. Quantitative analysis allowed for better discrimination and more accurate evaluation of low-intensity staining. In conclusion, AMACR protein is expressed in normal breast and its expression seems to increase in invasive carcinomas. We observed stronger AMACR protein expression in high-grade carcinomas when compared with low-grade ones. Quantitative image analysis is a novel way to accurately and reproducibly evaluate immunohistochemistry in breast tissue samples using high-density tissue microarrays.     

Role of a second stage partogram in predicting the outcome of normal labour

Background: Management of the second stage of labour is dictated by arbitrary time limits rather than true measures of progress. No partogram is available for second stage of labour.
Objectives: To evaluate a partogram designed for use for the second stage of labour.
Methods: This prospective cross-sectional analytical study included low-risk pregnant women with singleton fetuses with vertex presentations at term. From onset of the second stage, vaginal examinations were performed every 30 min until delivery. A scoring system developed by Sizer et al . was used based on station and position of fetal head. Scores were plotted on a second stage partogram and used to predict labour outcomes, such as duration of second stage and mode of delivery.
Results: Of 79 women examined, 73 had spontaneous vaginal delivery. Of the remaining six, four required oxytocin infusion and other two required vacuum extraction. The median durations of the second stage of labour for primigravidas ( n  = 34) and multigravidas ( n  = 45) were 35 and 25 min, respectively. The median Sizer's partogram score at the onset of second stage was 4. Multiple regression analysis showed that the partogram score ( r ² = 0.27) and gravidity ( r ² = 0.10) were independent predictors of duration of the second stage. There was a significant association between second stage progress plotted to the right of the partogram line and non-spontaneous delivery ( P =  0.01).
Conclusion: The second stage partogram score at onset can predict the duration of second stage. Poor progress plotted on the partogram is associated with non-spontaneous delivery.     

Chemokine-induced leishmanicidal activity in murine macrophages via the generation of nitric oxide

This study explored the role of the proinflammatory chemokines macrophage inflammatory protein (MIP)-1alpha and macrophage chemoattractant protein (MCP)-1 for development of antileishmanial activity. There was substantial inhibition in nitrite generation in Leishmania donovani-infected macrophages. A marked elevation of nitrite generation and induction of inducible nitric oxide (NO) synthase (iNOS) mRNA was found in chemokine-primed parasite-infected macrophages. Tumor necrosis factor-alpha, which is the priming signal for NO production, was also up-regulated under similar experimental conditions. The priming with chemokine inhibited the multiplication of L. donovani amastigotes within the intramacrophageal milieu. The antileishmanial effect of chemokines was almost completely abrogated when the macrophages were preincubated with l-N-monomethyl arginine, the specific inhibitor of iNOS. The results of this investigation suggest that the CC chemokines MIP-1alpha and MCP-1 orchestrate an antileishmanial armamentarium via the induction of an NO-mediated regulatory mechanism to control the intracellular growth and multiplication of the Leishmania protozoan.     

Ischemia Modified Albumin for the assessment of patients presenting to the emergency department with acute chest pain but normal or non-diagnostic 12-lead electrocardiograms and negative cardiac troponin T

BACKGROUND: The diagnosis of myocardial ischemia in patients with acute chest pain at rest but non-diagnostic electrocardiograms (ECG) is problematic. Ischemia Modified Albumin (IMA) is a new biochemical marker of ischemia, which may be useful to characterise acute coronary syndrome (ACS) patients. METHODS: We studied 131 patients (mean age 58.5 years; 95 male) presenting to the emergency department with symptoms suggestive of ACS but with normal or non-diagnostic ECGs. Cardiac troponin T (cTnT) and IMA were measured within 3 h of last chest pain episode. Based on hospital diagnostic test results, patients were classified as having ACS or non-ischemic chest pain (NICP), by two independent cardiologists unaware of IMA results. RESULTS: Mean IMA levels (U/ml) were higher in patients with ACS (98.3+/-11) compared to patients with NICP (85.5+/-15); p<0.0001. IMA levels >93.5 U/ml demonstrated a sensitivity and specificity of 75% for the diagnosis of ACS; area under the receiver operator characteristic curve 0.78 (95% CI: 0.70-0.85). If we applied the manufacturer cutoff point of 85 U/ml, the sensitivity of IMA increased to 90.6% with a specificity of 49.3% (negative predictive value=84.6%). In combination with cTnT (6-12 h) (>0.05 ng/ml), the sensitivity increased to 92.2%. After multivariate analysis, IMA levels >85 U/ml (odds ratio=14.6 [95% CI 4.4-48.4]; p<0.0001), age and prior myocardial infarction were independent predictors of ACS. CONCLUSION: IMA may be a useful biomarker for the identification of ACS in patients presenting with typical acute chest pain but normal or non-diagnostic ECGs.