The occurrence, distribution, and genetic relatedness of double‐stranded RNA (dsRNA) components from 36 isolates of Rhizoctonia solani belonging to nine anastomosis groups (AGs) were studied using electrophoretic analysis and RNA–RNA blot hybridization. DsRNA was consistently detected in all 36 isolates. The size of the dsRNA components varied considerably, ranging from 0·74 to 23 kb. Two thirds of isolates possessed different size dsRNA components. Only two of the isolates had small size dsRNA between 0·5 and 1·0 kb. The biotin‐labelled dsRNA probes provided the sensitivity and specificity required to study genetic relationships of dsRNA. As little as 10 pg of ‘hybridizing’ dsRNA could be detected using biotin‐labelled total dsRNA with no detectable nonspecific‐hybridization. Results from several dot‐spot as well as RNA–RNA gel hybridization experiments revealed considerable sequence heterogeneity among dsRNA components within each isolate or isolates from the same AG.

No abstract

No abstract

No abstract

Breast cancer still imposes significant healthcare burden on women worldwide. Mammography is currently the benchmark for screening, but mammography may not; a) detect all early stage breast cancers especially in young women with dense breast, b) unequivocally distinguish between benign disease and cancer and c) be beneficial to people in remote and underserved areas without mammograms. Biomarkers could complement mammography because of the potential to enable earlier detection, accurate differentiation between non-cancerous and cancerous breast lesions, and allow screening of a specimen like blood collected from a remote site. As a step towards developing a biomarker-based blood test for early detection of breast disease, we have used the branched-DNA multiplex technology to measure the levels of 24 candidate biomarkers in blood of fully informed and consented female donors classified into three broad categories; normal, with breast cancer (BC), and with benign breast disease (BD). To determine if the groups can be identified using the levels of the 24 candidate biomarkers we trained prediction models based on the data without background subtraction and applied the same training process on background subtracted data. Next we used an in-house developed software, to identify the optimal transformation and normalization methods. Due to the small sample size available for the current study, we generated 1,000 cases and 1,000 controls by simulating the original data via a distributional sampling model, and used the simulated data as the test validation set (TVS). Three prediction models were then applied to the original data and simulated TVS, namely: in-house GA optimized k of m, Random Forest (Salford System), Logistic Regression (SAS). Random Forest proved to be the best model for group prediction displaying sensitivities and specificities above 90%. This presentation will discuss the extent to which the levels of the biomarkers in blood allowed the identification of donors classified as “normal” and with “breast disease”. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 4624.