The L1/2 regularization approach for survival analysis in the accelerated failure time model

An Adaptive Transfer-Learning-Based Deep Cox Neural Network for Hepatocellular Carcinoma Prognosis Prediction

Background

Predicting hepatocellular carcinoma (HCC) prognosis is important for treatment selection, and it is increasingly interesting to predict prognosis through gene expression data. Currently, the prognosis remains of low accuracy due to the high dimension but small sample size of liver cancer omics data. In previous studies, a transfer learning strategy has been developed by pre-training models on similar cancer types and then fine-tuning the pre-trained models on the target dataset. However, transfer learning has limited performance since other cancer types are similar at different levels, and it is not trivial to balance the relations with different cancer types.

Methods

Here, we propose an adaptive transfer-learning-based deep Cox neural network (ATRCN), where cancers are represented by 12 phenotype and 10 genotype features, and suitable cancers were adaptively selected for model pre-training. In this way, the pre-trained model can learn valuable prior knowledge from other cancer types while reducing the biases.

Results

ATRCN chose pancreatic and stomach adenocarcinomas as the pre-training cancers, and the experiments indicated that our method improved the C-index of 3.8% by comparing with traditional transfer learning methods. The independent tests on three additional HCC datasets proved the robustness of our model. Based on the divided risk subgroups, we identified 10 HCC prognostic markers, including one new prognostic marker, TTC36. Further wet experiments indicated that TTC36 is associated with the progression of liver cancer cells.

Conclusion

These results proved that our proposed deep-learning-based method for HCC prognosis prediction is robust, accurate, and biologically meaningful.

A Three-Way Optimization Technique for Noise Robust Moving Object Detection Using Tensor Low-Rank Approximation, l1/2, and TTV Regularizations

The rising demand for surveillance systems naturally necessitates more efficient and noise robust moving object detection (MOD) systems from the captured video streams. Inspired by the challenges in MOD which are yet to be addressed properly, this paper proposes a new MOD scheme using l_1/2 regularization in the tensor framework. It takes advantage of the special features of tensor singular value decomposition ( t -SVD) along with regularizations using l_1/2 -norm with half thresholding operation and tensor total variation (TTV) to develop a noise robust MOD system with improved detection accuracy. While t -SVD exploits the spatio-temporal correlation of the video background, l_1/2 regularization provides noise robustness besides removing the sparser but discontinuous dynamic elements in the spatio-temporal direction. Moreover, TTV enhances the spatio-temporal continuity and fills up the gaps due to the lingering objects and thereby extracting the foreground precisely. The proposed three-way optimization method is designed to address both static and dynamic background cases of MOD separately with the intention to reduce the misclassifications due to moving/cluttered background. The brilliance of this method is confirmed by the impressive visual quality of the background/foreground separation, noise robustness, reduced computational complexity, and rapid response. The quantitative evaluation discloses the predominance of the proposed method with respect to the state-of-the-art techniques.

Biomarker discovery for predicting spontaneous preterm birth from gene expression data by regularized logistic regression

In this work, we provide a computational method of regularized logistic regression for discovering biomarkers of spontaneous preterm birth (SPTB) from gene expression data. The successful identification of SPTB biomarkers will greatly benefit the interference of infant gestational age for reducing the risks of pregnant women and preemies. In recent years, various approaches have been proposed for the feature selection of identifying the subset of meaningful genes that can achieve accurate classification for disease samples from controls. Here, we comprehensively summarize the regularized logistic regression with seven effective penalties developed for the selection of strongly indicative genes of SPTB from microarray data. We compare their properties and assess their classification performances in multiple datasets. It shows that elastic net, lasso,L 1 / 2 and SCAD penalties get the better performance than others and can be successfully used to identify biomarkers of SPTB. Particularly, we make a functional enrichment analysis on these biomarkers and construct a logistic regression classifier based on them. The classifier generates an indicator of preterm risk score (PRS) for predicting SPTB. Based on the trained predictor, we verify the identified biomarkers on an independent dataset. The biomarkers achieve the AUC value of 0.933 in the SPTB classification. The results demonstrate the effectiveness and efficiency of the built-up strategy of biomarker discovery with regularized logistic regression. Obviously, the proposed method of discovering biomarkers for SPTB can be easily extended for other complex diseases.

Deep learning-based ovarian cancer subtypes identification using multi-omics data

Background

Identifying molecular subtypes of ovarian cancer is important. Compared to identify subtypes using single omics data, the multi-omics data analysis can utilize more information. Autoencoder has been widely used to construct lower dimensional representation for multi-omics feature integration. However, learning in the deep architectures in Autoencoder is difficult for achieving satisfied generalization performance. To solve this problem, we proposed a novel deep learning-based framework to robustly identify ovarian cancer subtypes by using denoising Autoencoder.

Results

In proposed method, the composite features of multi-omics data in the Cancer Genome Atlas were produced by denoising Autoencoder, and then the generated low-dimensional features were input into k-means for clustering. At last based on the clustering results, we built the light-weighted classification model with L1-penalized logistic regression method. Furthermore, we applied the differential expression analysis and WGCNA analysis to select target genes related to molecular subtypes. We identified 34 biomarkers and 19 KEGG pathways associated with ovarian cancer.

Conclusions

The independent test results in three GEO datasets proved the robustness of our model. The literature reviewing show 19 (56%) biomarkers and 8(42.1%) KEGG pathways identified based on the classification subtypes have been proved to be associated with ovarian cancer. The outcomes indicate that our proposed method is feasible and can provide reliable results.

Structured sparsity regularization for analyzing high-dimensional omics data

The development of new molecular and cell technologies is having a significant impact on the quantity of data generated nowadays. The growth of omics databases is creating a considerable potential for knowledge discovery and, concomitantly, is bringing new challenges to statistical learning and computational biology for health applications. Indeed, the high dimensionality of these data may hamper the use of traditional regression methods and parameter estimation algorithms due to the intrinsic non-identifiability of the inherent optimization problem. Regularized optimization has been rising as a promising and useful strategy to solve these ill-posed problems by imposing additional constraints in the solution parameter space. In particular, the field of statistical learning with sparsity has been significantly contributing to building accurate models that also bring interpretability to biological observations and phenomena. Beyond the now-classic elastic net, one of the best-known methods that combine lasso with ridge penalizations, we briefly overview recent literature on structured regularizers and penalty functions that have been applied in biomedical data to build parsimonious models in a variety of underlying contexts, from survival to generalized linear models. These methods include functions of $\ell _k$-norms and network-based penalties that take into account the inherent relationships between the features. The successful application to omics data illustrates the potential of sparse structured regularization for identifying disease's molecular signatures and for creating high-performance clinical decision support systems towards more personalized healthcare. Supplementary information: Supplementary data are available at Briefings in Bioinformatics online.

A novel Log penalty in a path seeking scheme for biomarker selection

Biomarker selection or feature selection from survival data is a topic of considerable interest. Recently various survival analysis approaches for biomarker selection have been developed; however, there are growing challenges to currently methods for handling high-dimensional and low-sample problem. We propose a novel Log-sum regularization estimator within accelerated failure time (AFT) for predicting cancer patient survival time with a few biomarkers. This approach is implemented in path seeking algorithm to speed up solving the Log-sum penalty. Additionally, the control parameter of Log-sum penalty is modified by Bayesian information criterion (BIC). The results indicate that our proposed approach is able to achieve good performance in both simulated and real datasets with other ℓ1 type regularization methods for biomarker selection.

Robust sparse accelerated failure time model for survival analysis

To identify the bio-mark genes related to disease with high dimension and low sample size gene expression data, various regression approaches with different regularization methods have been proposed to solve this problem. Nevertheless, high-noises in biological data significantly reduce the performances of methods. The accelerated failure time (AFT) modelwas designed for gene selection and survival time estimation in cancer survival analysis. In this article, we proposed a novel robust sparse accelerated failure time model (RS-AFT) through combining the least absolute deviation (LAD) and Lq regularization. An iterative weighted linear programming algorithm without regularization parameter tuning was proposed to solve this RS-AFT model. The results of the experiments show our method has better performancebothin gene selection and survival time estimationthan some widely used regularization methods such as lasso, elastic net and SCAD. Hence we thought the RS-AFT model may be a competitive regularization method in cancer survival analysis.