Gene selection is crucial for cancer classification using microarray data. In the interests of improving cancer classification accuracy, in this paper, we developed a new wrapper method called ieGENES for gene selection. First we proposed a parsimonious kernel machine regularization (PKMR) model by using ridge regularization in kernel machine driven classification to tackle multi-collinearity for the sake of stable estimates in high-dimensional settings. Then the ieGENES algorithm was developed to optimally identify relevant genes while iteratively eliminating redundant ones based on leave-one-out cross-validation accuracy. In particular, we developed a new methodology to optimally update model parameters upon gene removal. The ieGENES algorithm was evaluated on six cancer microarray datasets and compared to existing methods. Classification accuracy and number of differentially expressed genes (DEGs) identified were assessed. In terms of gene selection accuracy, the ieGENESoutperformed multiple wrapper methods on 5 out of 6 datasets (Colon, Leukemia, Hepato, Glioma, and Breast Cancers), with statistically significant improvements (p<0.001). For the Colon dataset, ieGENES achieved 96.21% accuracy with 167 DEGs. The proposed ieGENES technique demonstrated superior performance in identifying DEGs for cancer diagnosis comparing with existing techniques. It offers a promising tool for identifying biologically relevant genes in microarray data analysis and biomarker discovery for cancer research.

Multiple transcriptomic predictors of tumour cell radiosensitivity (RS) have been proposed, but they have not been benchmarked against one another or to control models. To address this, we present RadSigBench, a comprehensive benchmarking framework for RS signatures. The approach compares candidate models to those developed from randomly resampled control signatures and from cellular processes integral to the radiation response. Robust evaluation of signature accuracy, both overall and for individual tissues, is performed. The NCI60 and Cancer Cell Line Encyclopaedia datasets are integrated into our workflow. Prediction of two measures of RS is assessed: survival fraction after 2 Gy and mean inactivation dose. We apply the RadSigBench framework to seven prominent published signatures of radiation sensitivity and test for equivalence to control signatures. The mean out-of-sample R2 for the published models on test data was very poor at 0.01 (range: -0.05 to 0.09) for Cancer Cell Line Encyclopedia and 0.00 (range: -0.19 to 0.19) in the NCI60 data. The accuracy of both published and cellular process signatures investigated was equivalent to the resampled controls, suggesting that these signatures contain limited radiation-specific information. Enhanced modelling strategies are needed for effective prediction of intrinsic RS to inform clinical treatment regimes. We make recommendations for methodological improvements, for example the inclusion of perturbation data, multiomics, advanced machine learning and mechanistic modelling. Our validation framework provides for robust performance assessment of ongoing developments in intrinsic RS prediction.

The ability to rank proteins by their likely success in crystallization is useful in current Structural Biology efforts and in particular in high-throughput Structural Genomics initiatives. We present ParCrys, a Parzen Window approach to estimate a protein's propensity to produce diffraction-quality crystals. The Protein Data Bank (PDB) provided training data whilst the databases TargetDB and PepcDB were used to define feature selection data as well as test data independent of feature selection and training. ParCrys outperforms the OB-Score, SECRET and CRYSTALP on the data examined, with accuracy and Matthews correlation coefficient values of 79.1% and 0.582, respectively (74.0% and 0.227, respectively, on data with a 'real-world' ratio of positive:negative examples). ParCrys predictions and associated data are available from www.compbio.dundee.ac.uk/parcrys.