Deep learning is a promising technology to accurately select individuals with high phenotypic values based on genotypic data. Genomic selection (GS) is a promising breeding strategy by which the phenotypes of plant individuals are usually predicted based on genome-wide markers of genotypes. In this study, we present a deep learning method, named DeepGS, to predict phenotypes from genotypes. Using a deep convolutional neural network, DeepGS uses hidden variables that jointly represent features in genotypes when making predictions; it also employs convolution, sampling and dropout strategies to reduce the complexity of high-dimensional genotypic data. We used a large GS dataset to train DeepGS and compared its performance with other methods. The experimental results indicate that DeepGS can be used as a complement to the commonly used RR-BLUP in the prediction of phenotypes from genotypes. The complementarity between DeepGS and RR-BLUP can be utilized using an ensemble learning approach for more accurately selecting individuals with high phenotypic values, even for the absence of outlier individuals and subsets of genotypic markers. The source codes of DeepGS and the ensemble learning approach have been packaged into Docker images for facilitating their applications in different GS programs.

The emergence of epitranscriptome opened a new chapter in gene regulation. 5-methylcytosine (m5C), as an important post-transcriptional modification, has been identified to be involved in a variety of biological processes such as subcellular localization and translational fidelity. Though high-throughput experimental technologies have been developed and applied to profile m5C modifications under certain conditions, transcriptome-wide studies of m5C modifications are still hindered by the dynamic nature of m5C and the lack of computational prediction methods. In this study, we introduced PEA-m5C, a machine learning-based m5C predictor trained with features extracted from the flanking sequence of m5C modifications. PEA-m5C yielded an average AUC (area under the receiver operating characteristic) of 0.939 in 10-fold cross-validation experiments based on known Arabidopsis m5C modifications. A rigorous independent testing showed that PEA-m5C (Accuracy [Acc] = 0.835, Matthews correlation coefficient [MCC] = 0.688) is remarkably superior to the recently developed m5C predictor iRNAm5C-PseDNC (Acc = 0.665, MCC = 0.332). PEA-m5C has been applied to predict candidate m5C modifications in annotated Arabidopsis transcripts. Further analysis of these m5C candidates showed that 4nt downstream of the translational start site is the most frequently methylated position. PEA-m5C is freely available to academic users at: https://github.com/cma2015/PEA-m5C.

Supplementary data are available at Bioinformatics online.

These results indicate that perceptual learning treatment for amblyopia had a positive effect on the visual cortex and temporal lobe visual areas in patients with anisometropic amblyopia.