Recursive Random Lasso (RRLasso) for Identifying Anti-Cancer Drug Targets

VSOLassoBag: a variable-selection oriented LASSO bagging algorithm for biomarker discovery in omic-based translational research

Screening biomolecular markers from high-dimensional biological data is one of the long-standing tasks for biomedical translational research. With its advantages in both feature shrinkage and biological interpretability, Least Absolute Shrinkage and Selection Operator (LASSO) algorithm is one of the most popular methods for the scenarios of clinical biomarker development. However, in practice, applying LASSO on omics-based data with high dimensions and low-sample size may usually result in an excess number of predictive variables, leading to the overfitting of the model. Here, we present VSOLassoBag, a wrapped LASSO approach by integrating an ensemble learning strategy to help select efficient and stable variables with high confidence from omics-based data. Using a bagging strategy in combination with a parametric method or inflection point search method, VSOLassoBag can integrate and vote variables generated from multiple LASSO models to determine the optimal candidates. The application of VSOLassoBag on both simulation datasets and real-world datasets shows that the algorithm can effectively identify markers for either case-control binary classification or prognosis prediction. In addition, by comparing with multiple existing algorithms, VSOLassoBag shows a comparable performance under different scenarios while resulting in fewer features than others. In summary, VSOLassoBag, which is available at https://seqworld.com/VSOLassoBag/ under the GPL v3 license, provides an alternative strategy for selecting reliable biomarkers from high-dimensional omics data. For user's convenience, we implement VSOLassoBag as an R package that provides multithreading computing configurations.

Prediction of drug sensitivity based on multi-omics data using deep learning and similarity network fusion approaches

With the rapid development of multi-omics technologies and accumulation of large-scale bio-datasets, many studies have conducted a more comprehensive understanding of human diseases and drug sensitivity from multiple biomolecules, such as DNA, RNA, proteins and metabolites. Using single omics data is difficult to systematically and comprehensively analyze the complex disease pathology and drug pharmacology. The molecularly targeted therapy-based approaches face some challenges, such as insufficient target gene labeling ability, and no clear targets for non-specific chemotherapeutic drugs. Consequently, the integrated analysis of multi-omics data has become a new direction for scientists to explore the mechanism of disease and drug. However, the available drug sensitivity prediction models based on multi-omics data still have problems such as overfitting, lack of interpretability, difficulties in integrating heterogeneous data, and the prediction accuracy needs to be improved. In this paper, we proposed a novel drug sensitivity prediction (NDSP) model based on deep learning and similarity network fusion approaches, which extracts drug targets using an improved sparse principal component analysis (SPCA) method for each omics data, and construct sample similarity networks based on the sparse feature matrices. Furthermore, the fused similarity networks are put into a deep neural network for training, which greatly reduces the data dimensionality and weakens the risk of overfitting problem. We use three omics of data, RNA sequence, copy number aberration and methylation, and select 35 drugs from Genomics of Drug Sensitivity in Cancer (GDSC) for experiments, including Food and Drug Administration (FDA)-approved targeted drugs, FDA-unapproved targeted drugs and non-specific therapies. Compared with some current deep learning methods, our proposed method can extract highly interpretable biological features to achieve highly accurate sensitivity prediction of targeted and non-specific cancer drugs, which is beneficial for the development of precision oncology beyond targeted therapy.

Hi-LASSO: High-performance python and apache spark packages for feature selection with high-dimensional data

High-dimensional LASSO (Hi-LASSO) is a powerful feature selection tool for high-dimensional data. Our previous study showed that Hi-LASSO outperformed the other state-of-the-art LASSO methods. However, the substantial cost of bootstrapping and the lack of experiments for a parametric statistical test for feature selection have impeded to apply Hi-LASSO for practical applications. In this paper, the Python package and its Spark library are efficiently designed in a parallel manner for practice with real-world problems, as well as providing the capability of the parametric statistical tests for feature selection on high-dimensional data. We demonstrate Hi-LASSO's outperformance with various intensive experiments in a practical manner. Hi-LASSO will be efficiently and easily performed by using the packages for feature selection. Hi-LASSO packages are publicly available at https://github.com/datax-lab/Hi-LASSO under the MIT license. The packages can be easily installed by Python PIP, and additional documentation is available at https://pypi.org/project/hi-lasso and https://pypi.org/project/Hi-LASSO-spark.

Selecting Genetic Variants and Interactions Associated with Amyotrophic Lateral Sclerosis: A Group LASSO Approach

Amyotrophic lateral sclerosis (ALS) is a multi-system neurodegenerative disease that affects both upper and lower motor neurons, resulting from a combination of genetic, environmental, and lifestyle factors. Usually, the association between single-nucleotide polymorphisms (SNPs) and this disease is tested individually, which leads to the testing of multiple hypotheses. In addition, this classical approach does not support the detection of interaction-dependent SNPs. We applied a two-step procedure to select SNPs and pairwise interactions associated with ALS. SNP data from 276 ALS patients and 268 controls were analyzed by a two-step group LASSO in 2000 iterations. In the first step, we fitted a group LASSO model to a bootstrap sample and a random subset of predictors (25%) from the original data set aiming to screen for important SNPs and, in the second step, we fitted a hierarchical group LASSO model to evaluate pairwise interactions. An in silico analysis was performed on a set of variables, which were prioritized according to their bootstrap selection frequency. We identified seven SNPs (rs16984239, rs10459680, rs1436918, rs1037666, rs4552942, rs10773543, and rs2241493) and two pairwise interactions (rs16984239:rs2118657 and rs16984239:rs3172469) potentially involved in nervous system conservation and function. These results may contribute to the understanding of ALS pathogenesis, its diagnosis, and therapeutic strategy improvement.

A Review of Current In Silico Methods for Repositioning Drugs and Chemical Compounds

Drug repositioning is a new way of applying the existing therapeutics to new disease indications. Due to the exorbitant cost and high failure rate in developing new drugs, the continued use of existing drugs for treatment, especially anti-tumor drugs, has become a widespread practice. With the assistance of high-throughput sequencing techniques, many efficient methods have been proposed and applied in drug repositioning and individualized tumor treatment. Current computational methods for repositioning drugs and chemical compounds can be divided into four categories: (i) feature-based methods, (ii) matrix decomposition-based methods, (iii) network-based methods, and (iv) reverse transcriptome-based methods. In this article, we comprehensively review the widely used methods in the above four categories. Finally, we summarize the advantages and disadvantages of these methods and indicate future directions for more sensitive computational drug repositioning methods and individualized tumor treatment, which are critical for further experimental validation.

Evaluating the consistency of large-scale pharmacogenomic studies

Recent years have seen an increase in the availability of pharmacogenomic databases such as Genomics of Drug Sensitivity in Cancer (GDSC) and Cancer Cell Line Encyclopedia (CCLE) that provide genomic and functional characterization information for multiple cell lines. Studies have alluded to the fact that specific characterizations may be inconsistent between different databases. Analysis of the potential discrepancies in the different databases is highly significant, as these sources are frequently used to analyze and validate methodologies for personalized cancer therapies. In this article, we review the recent developments in investigating the correspondence between different pharmacogenomics databases and discuss the potential factors that require attention when incorporating these sources in any modeling analysis. Furthermore, we explored the consistency among these databases using copulas that can capture nonlinear dependencies between two sets of data.

Network-based Biased Tree Ensembles (NetBiTE) for Drug Sensitivity Prediction and Drug Sensitivity Biomarker Identification in Cancer

We present the Network-based Biased Tree Ensembles (NetBiTE) method for drug sensitivity prediction and drug sensitivity biomarker identification in cancer using a combination of prior knowledge and gene expression data. Our devised method consists of a biased tree ensemble that is built according to a probabilistic bias weight distribution. The bias weight distribution is obtained from the assignment of high weights to the drug targets and propagating the assigned weights over a protein-protein interaction network such as STRING. The propagation of weights, defines neighborhoods of influence around the drug targets and as such simulates the spread of perturbations within the cell, following drug administration. Using a synthetic dataset, we showcase how application of biased tree ensembles (BiTE) results in significant accuracy gains at a much lower computational cost compared to the unbiased random forests (RF) algorithm. We then apply NetBiTE to the Genomics of Drug Sensitivity in Cancer (GDSC) dataset and demonstrate that NetBiTE outperforms RF in predicting IC50 drug sensitivity, only for drugs that target membrane receptor pathways (MRPs): RTK, EGFR and IGFR signaling pathways. We propose based on the NetBiTE results, that for drugs that inhibit MRPs, the expression of target genes prior to drug administration is a biomarker for IC50 drug sensitivity following drug administration. We further verify and reinforce this proposition through control studies on, PI3K/MTOR signaling pathway inhibitors, a drug category that does not target MRPs, and through assignment of dummy targets to MRP inhibiting drugs and investigating the variation in NetBiTE accuracy.

Robust Sample-Specific Stability Selection with Effective Error Control

Identifying individual characteristics is a crucial issue in personalized genome research. To effectively identify sample-specific characteristics, we propose a novel strategy called robust sample-specific stability selection. Although stability selection shows effective feature selection results and has attractive theoretical property (i.e., per-family error rate control), the method's results are sensitive to the value of the regularization parameter because the method performs feature selection based only on the particular parameter value that maximizes the selection probability. To resolve this issue, we propose robust stability selection and show that our method provides an effective theoretical property (i.e., effective per-family error rate control). We also propose a sample-specific random lasso based on the kernel-based L₁-type regularization and weighted random sampling. The proposed robust sample-specific stability selection estimates the selection probabilities of variables using the sample-specific random lasso and then selects variables based on robust stability selection. Our method controls the effect of samples on sample-specific analysis by the two-stage strategy (i.e., the weighted random sampling and the kernel-based L₁-type approach in the random lasso), and thus we can effectively perform sample-specific analysis without disturbances of samples having characteristics different from those of the target sample. We observe from the numerical studies that our strategies can effectively perform sample-specific analysis and provide biologically reliable results in gene selection.

Identification of risk factors associated with carriage of resistant Escherichia coli in three culturally diverse ethnic groups in Tanzania: a biological and socioeconomic analysis

BACKGOUND

Improved antimicrobial stewardship, sanitation, and hygiene are WHO-inspired priorities for restriction of the spread of antimicrobial resistance. Prioritisation among these objectives is essential, particularly in low-income and middle-income countries, but the factors contributing most to antimicrobial resistance are typically unknown and could vary substantially between and within countries. We aimed to identify the biological and socioeconomic risk factors associated with carriage of resistant Escherichia coli in three culturally diverse ethnic groups in northern Tanzania.

METHODS

We developed a survey containing more than 200 items and administered it in randomly selected households in 13 Chagga, Arusha, or Maasai villages chosen on the basis of ethnic composition and distance to urban centres. Human stool samples were collected from a subset of households, as were liquid milk samples and swabs of milk containers. Samples were processed and plated onto MacConkey agar plates, then presumptive E coli isolates were identified on the basis of colony morphology. Susceptibility of isolates was then tested against a panel of nine antimicrobials (ampicillin, ceftazidime, chloramphenicol, ciprofloxacin, kanamycin, streptomycin, sulfamethoxazole, tetracycline, and trimethoprim) via a breakpoint assay. Susceptibility findings were matched with data across a wide range of household characteristics, including education, hygiene practices, wealth, livestock husbandry, and antibiotic use.

FINDINGS

Between March 23, 2012, and July 30, 2015, we interviewed 391 households (118 Arusha, 100 Chagga, and 173 Maasai). Human stool samples were collected at 226 (58%) households across the 13 villages. 181 milk samples and 191 milk-container swabs were collected from 117 households across seven villages. 11 470 putative E coli samples were isolated from stool samples. Antimicrobial use in people and livestock was not associated with prevalence of resistance at the household level. Instead, the factors with the greatest predictive value involved exposure to bacteria, and were intimately connected with fundamental cultural differences across study groups. These factors included how different subsistence types (pastoralists vs farmers) access water sources and consumption of unboiled milk, reflecting increased exposure to resistant bacteria in milk.

INTERPRETATION

When cultural and ecological conditions favour bacterial transmission, there is a high likelihood that people will harbour antimicrobial-resistant bacteria irrespective of antimicrobial use practices. Public health interventions to limit antimicrobial resistance need to be tailored to local practices that affect bacterial transmission.

FUNDING

US National Science Foundation; Biotechnology and Biological Sciences Research Council, UK Medical Research Council; and the Allen School.

Interaction-Based Feature Selection for Uncovering Cancer Driver Genes Through Copy Number-Driven Expression Level

Driver gene selection is crucial to understand the heterogeneous system of cancer. To identity cancer driver genes, various statistical strategies have been proposed, especially the L₁-type regularization methods have drawn a large amount of attention. However, the statistical approaches have been developed purely from algorithmic and statistical point, and the existing studies have applied the statistical approaches to genomic data analysis without consideration of biological knowledge. We consider a statistical strategy incorporating biological knowledge to identify cancer driver gene. The alterations of copy number have been considered to driver cancer pathogenesis processes, and the region of strong interaction of copy number alterations and expression levels was known as a tumor-related symptom. We incorporate the influence of copy number alterations on expression levels to cancer driver gene-selection processes. To quantify the dependence of copy number alterations on expression levels, we consider [Formula: see text] and [Formula: see text] effects of copy number alterations on expression levels of genes, and incorporate the symptom of tumor pathogenesis to gene-selection procedures. We then proposed an interaction-based feature-selection strategy based on the adaptive L₁-type regularization and random lasso procedures. The proposed method imposes a large amount of penalty on genes corresponding to a low dependency of the two features, thus the coefficients of the genes are estimated to be small or exactly 0. It implies that the proposed method can provide biologically relevant results in cancer driver gene selection. Monte Carlo simulations and analysis of the Cancer Genome Atlas (TCGA) data show that the proposed strategy is effective for high-dimensional genomic data analysis. Furthermore, the proposed method provides reliable and biologically relevant results for cancer driver gene selection in TCGA data analysis.