Supplementary data are available at Bioinformatics online.

Large-scale "omics" data have been increasingly used as an important resource for prognostic prediction of diseases and detection of associated genes. However, there are considerable challenges in analyzing high-dimensional molecular data, including the large number of potential molecular predictors, limited number of samples, and small effect of each predictor. We propose new Bayesian hierarchical generalized linear models, called spike-and-slab lasso GLMs, for prognostic prediction and detection of associated genes using large-scale molecular data. The proposed model employs a spike-and-slab mixture double-exponential prior for coefficients that can induce weak shrinkage on large coefficients, and strong shrinkage on irrelevant coefficients. We have developed a fast and stable algorithm to fit large-scale hierarchal GLMs by incorporating expectation-maximization (EM) steps into the fast cyclic coordinate descent algorithm. The proposed approach integrates nice features of two popular methods, i.e., penalized lasso and Bayesian spike-and-slab variable selection. The performance of the proposed method is assessed via extensive simulation studies. The results show that the proposed approach can provide not only more accurate estimates of the parameters, but also better prediction. We demonstrate the proposed procedure on two cancer data sets: a well-known breast cancer data set consisting of 295 tumors, and expression data of 4919 genes; and the ovarian cancer data set from TCGA with 362 tumors, and expression data of 5336 genes. Our analyses show that the proposed procedure can generate powerful models for predicting outcomes and detecting associated genes. The methods have been implemented in a freely available R package BhGLM

Supplementary data are available at Bioinformatics online.

OBJECTIVE The optimal margin size in postoperative stereotactic radiosurgery (SRS) for brain metastases is unknown. Herein, the authors investigated the effect of SRS planning target volume (PTV) margin on local recurrence and symptomatic radiation necrosis postoperatively. METHODS Records of patients who received postoperative LINAC-based SRS for brain metastases between 2006 and 2016 were reviewed and stratified based on PTV margin size (1.0 or > 1.0 mm). Patients were treated using frameless and framed SRS techniques, and both single-fraction and hypofractionated dosing were used based on lesion size. Kaplan-Meier and cumulative incidence models were used to estimate survival and intracranial outcomes, respectively. Multivariate analyses were also performed. RESULTS A total of 133 patients with 139 cavities were identified; 36 patients (27.1%) and 35 lesions (25.2%) were in the 1.0-mm group, and 97 patients (72.9%) and 104 lesions (74.8%) were in the > 1.0-mm group. Patient characteristics were balanced, except the 1.0-mm cohort had a better Eastern Cooperative Group Performance Status (grade 0: 36.1% vs 19.6%), higher mean number of brain metastases (1.75 vs 1.31), lower prescription isodose line (80% vs 95%), and lower median single fraction-equivalent dose (15.0 vs 17.5 Gy) (all p < 0.05). The median survival and follow-up for all patients were 15.6 months and 17.7 months, respectively. No significant difference in local recurrence was noted between the cohorts. An increased 1-year rate of symptomatic radionecrosis was seen in the larger margin group (20.9% vs 6.0%, p = 0.028). On multivariate analyses, margin size > 1.0 mm was associated with an increased risk for symptomatic radionecrosis (HR 3.07, 95% CI 1.13-8.34; p = 0.028), while multifraction SRS emerged as a protective factor for symptomatic radionecrosis (HR 0.13, 95% CI 0.02-0.76; p = 0.023). CONCLUSIONS Expanding the PTV margin beyond 1.0 mm is not associated with improved local recurrence but appears to increase the risk of symptomatic radionecrosis after postoperative SRS.

The package is freely available from the public GitHub repository, https://github.com/nyiuab/BhGLM.