The superior performance of machine-learning scoring functions for docking has caused a series of debates on whether it is due to learning knowledge from training data that are similar in some sense to the test data. With a systematically revised methodology and a blind benchmark realistically mimicking the process of prospective prediction of binding affinity, we have evaluated three broadly used classical scoring functions and five machine-learning counterparts calibrated with both random forest and extreme gradient boosting using both solo and hybrid features, showing for the first time that machine-learning scoring functions trained exclusively on a proportion of as low as 8% complexes dissimilar to the test set already outperform classical scoring functions, a percentage that is far lower than what has been recently reported on all the three CASF benchmarks. The performance of machine-learning scoring functions is underestimated due to the absence of similar samples in some artificially created training sets that discard the full spectrum of complexes to be found in a prospective environment. Given the inevitability of any degree of similarity contained in a large dataset, the criteria for scoring function selection depend on which one can make the best use of all available materials. Software code and data are provided at https://github.com/cusdulab/MLSF for interested readers to rapidly rebuild the scoring functions and reproduce our results, even to make extended analyses on their own benchmarks.

Abstract Purpose Accurate and early detection of breast cancer using effective imaging modalities is an active area of research in medical image analysis. Computeraided diagnosis (CAD) of breast cancer using digital mammograms may help in early diagnosis and can assist in maintaining patient health. The breast imaging reporting and data system (BIRADS) is widely used for risk assessment and classification grading in breast cancer screening. It contains seven different grading systems for breast cancer risk assessment. These range from grade 0 (incomplete) to grade 6 (proven malignancy). All other intermediate stages state the progression of risk. Methods The current research results have shown that two‐dimensional synthesized mammogram (2DSM) imaging and conventional full‐field digital mammography (FFDM) are two important imaging modalities which can be used for screening breast cancer. To the best of our knowledge, there is no study which has yet compared the BIRADS discrimination power of 2DSM and FFDM imaging modalities. In this paper we present a novel CAD‐based comparative study, using texton and gist for the characterization of breast cancer with 2DSM and FFDM imagery. Results The developed method achieved an average performance of 92.9% accuracy using a probabilistic neural network classifier for FFDM images with tenfold crossvalidation. Hence, our proposed model showed that FFDM images are more effective than the 2DSM imaging modality in discriminating BIRADS grades. Conclusion The obtained results confirmed that our method performed well in the early detection of breast cancer. Consequently, it can be used as a distinct system in rural hospitals.

SUMMARY: Serial analysis of gene expression (SAGE) provides an alternative, with additional advantages, to microarray gene expression studies. GonadSAGE is the first publicly available web-based SAGE database on male gonad development that covers six male mouse embryonic gonad stages, including E10.5, E11.5, E12.5, E13.5, E15.5 and E17.5. The sequence coverage of each SAGE library is beyond 150K, 'which is the most extensive sequence-based male gonadal transcriptome to date'. An interactive web interface with customizable parameters is provided for analyzing male gonad transcriptome information. Furthermore, the data can be visualized and analyzed with the other genomic features in the UCSC genome browser. It represents an integrated platform that leads to a better understanding of male gonad development, and allows discovery of related novel targets and regulatory pathways.