Genome-wide association analysis by lasso penalized logistic regression

Multi-scale hierarchical brain regions detect individual and interspecies variations of structural connectivity in macaque monkeys and humans

Macaques are representative animal models in translational research. However, the distinct shape and location of the brain regions between macaques and humans prevents us from comparing the brain structure directly. Here, we calculated structural connectivity (SC) with multi-scale hierarchical regions of interest (ROIs) to parcel out human and macaque brain into 8 (level 1 ROIs), 28 (level 2 ROIs), or 46 (level 3 ROIs) regions, which consist of anatomically and functionally defined level 4 ROIs (around 100 parcellation of the brain). The SC with the level 1 ROIs showed lower individual and interspecies variation in macaques and humans. SC with level 2 and 3 ROIs shows that the several regions in frontal, temporal and parietal lobe show distinct connectivity between macaques and humans. Lateral frontal cortex, motor cortex and auditory cortex were shown to be important areas for interspecies differences. These results provide insights to use macaques as animal models for translational study.

Multicategory Survival Outcomes Classification via Overlapping Group Screening Process Based on Multinomial Logistic Regression Model With Application to TCGA Transcriptomic Data

Objectives

Under the classification of multicategory survival outcomes of cancer patients, it is crucial to identify biomarkers that affect specific outcome categories. The classification of multicategory survival outcomes from transcriptomic data has been thoroughly investigated in computational biology. Nevertheless, several challenges must be addressed, including the ultra-high-dimensional feature space, feature contamination, and data imbalance, all of which contribute to the instability of the diagnostic model. Furthermore, although most methods achieve accurate predicted performance for binary classification with high-dimensional transcriptomic data, their extension to multi-class classification is not straightforward.

Methods

We employ the One-versus-One strategy to transform multi-class classification into multiple binary classification, and utilize the overlapping group screening procedure with binary logistic regression to include pathway information for identifying important genes and gene-gene interactions for multicategory survival outcomes.

Results

A series of simulation studies are conducted to compare the classification accuracy of our proposed approach with some existing machine learning methods. In practical data applications, we utilize the random oversampling procedure to tackle class imbalance issues. We then apply the proposed method to analyze transcriptomic data from various cancers in The Cancer Genome Atlas, such as kidney renal papillary cell carcinoma, lung adenocarcinoma, and head and neck squamous cell carcinoma. Our aim is to establish an accurate microarray-based multicategory cancer diagnosis model. The numerical results illustrate that the new proposal effectively enhances cancer diagnosis compared to approaches that neglect pathway information.

Conclusions

We showcase the effectiveness of the proposed method in terms of class prediction accuracy through evaluations on simulated synthetic datasets as well as real dataset applications. We also identified the cancer-related gene-gene interaction biomarkers and reported the corresponding network structure. According to the identified major genes and gene-gene interactions, we can predict for each patient the probabilities that he/she belongs to each of the survival outcome classes.

Network medicine-based epistasis detection in complex diseases: ready for quantum computing

Most heritable diseases are polygenic. To comprehend the underlying genetic architecture, it is crucial to discover the clinically relevant epistatic interactions (EIs) between genomic single nucleotide polymorphisms (SNPs) (1-3). Existing statistical computational methods for EI detection are mostly limited to pairs of SNPs due to the combinatorial explosion of higher-order EIs. With NeEDL (network-based epistasis detection via local search), we leverage network medicine to inform the selection of EIs that are an order of magnitude more statistically significant compared to existing tools and consist, on average, of five SNPs. We further show that this computationally demanding task can be substantially accelerated once quantum computing hardware becomes available. We apply NeEDL to eight different diseases and discover genes (affected by EIs of SNPs) that are partly known to affect the disease, additionally, these results are reproducible across independent cohorts. EIs for these eight diseases can be interactively explored in the Epistasis Disease Atlas (https://epistasis-disease-atlas.com). In summary, NeEDL demonstrates the potential of seamlessly integrated quantum computing techniques to accelerate biomedical research. Our network medicine approach detects higher-order EIs with unprecedented statistical and biological evidence, yielding unique insights into polygenic diseases and providing a basis for the development of improved risk scores and combination therapies.

BLUPmrMLM: A Fast mrMLM Algorithm in Genome-wide Association Studies

Multilocus genome-wide association study has become the state-of-the-art tool for dissecting the genetic architecture of complex and multiomic traits. However, most existing multilocus methods require relatively long computational time when analyzing large datasets. To address this issue, in this study, we proposed a fast mrMLM method, namely, best linear unbiased prediction multilocus random-SNP-effect mixed linear model (BLUPmrMLM). First, genome-wide single-marker scanning in mrMLM was replaced by vectorized Wald tests based on the best linear unbiased prediction (BLUP) values of marker effects and their variances in BLUPmrMLM. Then, adaptive best subset selection (ABESS) was used to identify potentially associated markers on each chromosome to reduce computational time when estimating marker effects via empirical Bayes. Finally, shared memory and parallel computing schemes were used to reduce the computational time. In simulation studies, BLUPmrMLM outperformed GEMMA, EMMAX, mrMLM, and FarmCPU as well as the control method (BLUPmrMLM with ABESS removed), in terms of computational time, power, accuracy for estimating quantitative trait nucleotide positions and effects, false positive rate, false discovery rate, false negative rate, and F1 score. In the reanalysis of two large rice datasets, BLUPmrMLM significantly reduced the computational time and identified more previously reported genes, compared with the aforementioned methods. This study provides an excellent multilocus model method for the analysis of large-scale and multiomic datasets. The software mrMLM v5.1 is available at BioCode (https://ngdc.cncb.ac.cn/biocode/tool/BT007388) or GitHub (https://github.com/YuanmingZhang65/mrMLM).

Privacy-preserving model evaluation for logistic and linear regression using homomorphically encrypted genotype data

OBJECTIVE

Linear and logistic regression are widely used statistical techniques in population genetics for analyzing genetic data and uncovering patterns and associations in large genetic datasets, such as identifying genetic variations linked to specific diseases or traits. However, obtaining statistically significant results from these studies requires large amounts of sensitive genotype and phenotype information from thousands of patients, which raises privacy concerns. Although cryptographic techniques such as homomorphic encryption offers a potential solution to the privacy concerns as it allows computations on encrypted data, previous methods leveraging homomorphic encryption have not addressed the confidentiality of shared models, which can leak information about the training data.

METHODS

In this work, we present a secure model evaluation method for linear and logistic regression using homomorphic encryption for six prediction tasks, where input genotypes, output phenotypes, and model parameters are all encrypted.

RESULTS

Our method ensures no private information leakage during inference and achieves high accuracy (≥93% for all outcomes) with each inference taking less than ten seconds for ∼200 genomes.

CONCLUSION

Our study demonstrates that it is possible to perform linear and logistic regression model evaluation while protecting patient confidentiality with theoretical security guarantees. Our implementation and test data are available at https://github.com/G2Lab/privateML/.

Learning peptide properties with positive examples only

Deep learning can create accurate predictive models by exploiting existing large-scale experimental data, and guide the design of molecules. However, a major barrier is the requirement of both positive and negative examples in the classical supervised learning frameworks. Notably, most peptide databases come with missing information and low number of observations on negative examples, as such sequences are hard to obtain using high-throughput screening methods. To address this challenge, we solely exploit the limited known positive examples in a semi-supervised setting, and discover peptide sequences that are likely to map to certain antimicrobial properties via positive-unlabeled learning (PU). In particular, we use the two learning strategies of adapting base classifier and reliable negative identification to build deep learning models for inferring solubility, hemolysis, binding against SHP-2, and non-fouling activity of peptides, given their sequence. We evaluate the predictive performance of our PU learning method and show that by only using the positive data, it can achieve competitive performance when compared with the classical positive-negative (PN) classification approach, where there is access to both positive and negative examples.

Discovering significant topics from legal decisions with selective inference

We propose and evaluate an automated pipeline for discovering significant topics from legal decision texts by passing features synthesized with topic models through penalized regressions and post-selection significance tests. The method identifies case topics significantly correlated with outcomes, topic-word distributions which can be manually interpreted to gain insights about significant topics, and case-topic weights which can be used to identify representative cases for each topic. We demonstrate the method on a new dataset of domain name disputes and a canonical dataset of European Court of Human Rights violation cases. Topic models based on latent semantic analysis as well as language model embeddings are evaluated. We show that topics derived by the pipeline are consistent with legal doctrines in both areas and can be useful in other related legal analysis tasks. This article is part of the theme issue 'A complexity science approach to law and governance'.

The identification of signature genes and their relationship with immune cell infiltration in age-related macular degeneration

BACKGROUND

Age-related macular degeneration (AMD) is a prevalent source of visual impairment among the elderly population, and its incidence has risen in tandem with the increasing longevity of humans. Despite the progress made with anti-VEGF therapy, clinical outcomes have proven to be unsatisfactory.

METHOD

We obtained differentially expressed genes (DEGs) of AMD patients and healthy controls from the GEO database. GO and KEGG analyses were used to enrich the DEGs. Weighted gene coexpression network analysis (WGCNA) was used to identify modules related to AMD. SVM, random forest, and least absolute shrinkage and selection operator (LASSO) were employed to screen hub genes. Gene set enrichment analysis (GSEA) was used to explore the pathways in which these hub genes were enriched. CIBERSORT was utilized to analyze the relationship between the hub genes and immune cell infiltration. Finally, Western blotting and RT‒PCR were used to explore the expression of hub genes in AMD mice.

RESULTS

We screened 1084 DEGs in GSE29801, of which 496 genes were upregulated. These 1084 DEGs were introduced into the WGCNA, and 94 genes related to AMD were obtained. Seventy-nine overlapping genes were obtained by the Venn plot. These 79 genes were introduced into three machine-learning methods to screen the hub genes, and the genes identified by the three methods were TNC, FAP, SREBF1, and TGF-β2. We verified their diagnostic function in the GSE29801 and GSE103060 datasets. Then, the hub gene co-enrichment pathways were obtained by GO and KEGG analyses. CIBERSORT analysis showed that these hub genes were associated with immune cell infiltration. Finally, we found increased expression of TNC, FAP, SREBF1, and TGF-β2 mRNA and protein in the retinas of AMD mice.

CONCLUSION

We found that four hub genes, namely, FAP, TGF-β2, SREBF1, and TNC, have diagnostic significance in patients with AMD and are related to immune cell infiltration. Finally, we determined that the mRNA and protein expression of these hub genes was upregulated in the retinas of AMD mice.

Evaluation of penalized and machine learning methods for asthma disease prediction in the Korean Genome and Epidemiology Study (KoGES)

BACKGROUND

Genome-wide association studies have successfully identified genetic variants associated with human disease. Various statistical approaches based on penalized and machine learning methods have recently been proposed for disease prediction. In this study, we evaluated the performance of several such methods for predicting asthma using the Korean Chip (KORV1.1) from the Korean Genome and Epidemiology Study (KoGES).

RESULTS

First, single-nucleotide polymorphisms were selected via single-variant tests using logistic regression with the adjustment of several epidemiological factors. Next, we evaluated the following methods for disease prediction: ridge, least absolute shrinkage and selection operator, elastic net, smoothly clipped absolute deviation, support vector machine, random forest, boosting, bagging, naïve Bayes, and k-nearest neighbor. Finally, we compared their predictive performance based on the area under the curve of the receiver operating characteristic curves, precision, recall, F1-score, Cohen's Kappa, balanced accuracy, error rate, Matthews correlation coefficient, and area under the precision-recall curve. Additionally, three oversampling algorithms are used to deal with imbalance problems.

CONCLUSIONS

Our results show that penalized methods exhibit better predictive performance for asthma than that achieved via machine learning methods. On the other hand, in the oversampling study, randomforest and boosting methods overall showed better prediction performance than penalized methods.

Investigation of clear cell renal cell carcinoma grades using diffusion-relaxation correlation spectroscopic imaging with optimized spatial-spectrum analysis

OBJECTIVES

To differentiate high-grade from low-grade clear cell renal cell carcinoma (ccRCC) using diffusion-relaxation correlation spectroscopic imaging (DR-CSI) spectra in an equal separating analysis.

METHODS

Eighty patients with 86 pathologically confirmed ccRCCs who underwent DR-CSI were enrolled. Two radiologists delineated the region of interest. The spectrum was derived based on DR-CSI and was further segmented into multiple equal subregions from 2*2 to 9*9. The agreement between the 2 radiologists was assessed by the intraclass correlation coefficient (ICC). Logistic regression was used to establish the regression model for differentiation, and 5-fold cross-validation was used to evaluate its accuracy. McNemar's test was used to compare the diagnostic performance between equipartition models and the traditional parameters, including the apparent diffusion coefficient (ADC) and T2 value.

RESULTS

The inter-reader agreement decreased as the divisions in the equipartition model increased (overall ICC ranged from 0.859 to 0.920). The accuracy increased from the 2*2 to 9*9 equipartition model (0.68 for 2*2, 0.69 for 3*3 and 4*4, 0.70 for 5*5, 0.71 for 6*6, 0.78 for 7*7, and 0.75 for 8*8 and 9*9). The equipartition models with divisions >7*7 were significantly better than ADC and T2 (vs ADC: P = .002-.008; vs T2: P = .001-.004).

CONCLUSIONS

The equipartition method has the potential to analyse the DR-CSI spectrum and discriminate between low-grade and high-grade ccRCC.

ADVANCES IN KNOWLEDGE

The evaluation of DR-CSI relies on prior knowledge, and how to assess the spectrum derived from DR-CSI without prior knowledge has not been well studied.

Exploration of the pathophysiology of high myopia via proteomic profiling of human corneal stromal lenticules

This study aimed to investigate the underlying pathophysiology of high myopia by analyzing the proteome of human corneal stromal lenticule samples obtained through small incision lenticule extraction (SMILE). A total of thirty-two patients who underwent SMILE were included in the study. Label-free quantitative proteomic analysis was performed on corneal stromal lenticule samples, equally representing high myopia (n = 10) and low myopia (n = 10) groups. The identified and profiled lenticule proteomes were analyzed using in silico tools to explore biological characteristics of differentially expressed proteins (DEPs). Additionally, LASSO regression and random forest model were employed to identify key proteins associated with the pathophysiology of high myopia. The DEPs were found to be closely linked to immune activation, extracellular matrix, and cell adhesion-related pathways according to gene ontology analysis. Specifically, decreased expression of COL1A1 and increased expression of CDH11 were associated with the pathogenesis of high myopia and validated by western blotting (n = 6) and quantitative real time polymerase chain reaction (n = 6). Overall, this study provides evidence that COL1A1 and CDH11 may contribute to the pathophysiology of high myopia based on comparative proteomic profiling of human corneal stromal lenticules obtained through SMILE.

Pleiotropic genetic association analysis with multiple phenotypes using multivariate response best-subset selection

Genetic pleiotropy refers to the simultaneous association of a gene with multiple phenotypes. It is widely distributed in the whole genome and can help to understand the common genetic mechanism of diseases or traits. In this study, a multivariate response best-subset selection (MRBSS) model based pleiotropic association analysis method is proposed. Different from the traditional genetic association model, the high-dimensional genotypic data are viewed as response variables while the multiple phenotypic data as predictor variables. Moreover, the response best-subset selection procedure is converted into an 0-1 integer optimization problem by introducing a separation parameter and a tuning parameter. Furthermore, the model parameters are estimated by using the curve search under the modified Bayesian information criterion. Simulation experiments show that the proposed method MRBSS remarkably reduces the computational time, obtains higher statistical power under most of the considered scenarios, and controls the type I error rate at a low level. The application studies in the datasets of maize yield traits and pig lipid traits further verifies the effectiveness.

A human-interpretable machine learning pipeline based on ultrasound to support leiomyosarcoma diagnosis

The preoperative evaluation of myometrial tumors is essential to avoid delayed treatment and to establish the appropriate surgical approach. Specifically, the differential diagnosis of leiomyosarcoma (LMS) is particularly challenging due to the overlapping of clinical, laboratory and ultrasound features between fibroids and LMS. In this work, we present a human-interpretable machine learning (ML) pipeline to support the preoperative differential diagnosis of LMS from leiomyomas, based on both clinical data and gynecological ultrasound assessment of 68 patients (8 with LMS diagnosis). The pipeline provides the following novel contributions: (i) end-users have been involved both in the definition of the ML tasks and in the evaluation of the overall approach; (ii) clinical specialists get a full understanding of both the decision-making mechanisms of the ML algorithms and the impact of the features on each automatic decision. Moreover, the proposed pipeline addresses some of the problems concerning both the imbalance of the two classes by analyzing and selecting the best combination of the synthetic oversampling strategy of the minority class and the classification algorithm among different choices, and the explainability of the features at global and local levels. The results show very high performance of the best strategy (AUC = 0.99, F1 = 0.87) and the strong and stable impact of two ultrasound-based features (i.e., tumor borders and consistency of the lesions). Furthermore, the SHAP algorithm was exploited to quantify the impact of the features at the local level and a specific module was developed to provide a template-based natural language (NL) translation of the explanations for enhancing their interpretability and fostering the use of ML in the clinical setting.

Development and validation of a radiomics nomogram for diagnosis of malignant pleural effusion

OBJECTIVE

We aimed to develop a radiomics nomogram based on computed tomography (CT) scan features and high-throughput radiomics features for diagnosis of malignant pleural effusion (MPE).

METHODS

In this study, 507 eligible patients with PE (207 malignant and 300 benign) were collected retrospectively. Patients were divided into training (n = 355) and validation cohorts (n = 152). Radiomics features were extracted from initial unenhanced CT images. CT scan features of PE were also collected. We used the variance threshold algorithm and least absolute shrinkage and selection operator (LASSO) to select optimal features to build a radiomics model for predicting the nature of PE. Univariate and multivariable logistic regression analyzes were used to identify significant independent factors associated with MPE, which were then included in the radiomics nomogram.

RESULTS

A total of four CT features were retained as significant independent factors, including massive PE, obstructive atelectasis or pneumonia, pleural thickening > 10 mm, and pulmonary nodules and/or masses. The radiomics nomogram constructed from 13 radiomics parameters and four CT features showed good predictive efficacy in training cohort [area under the curve (AUC) = 0.926, 95% CI 0.894, 0.951] and validation cohort (AUC = 0.916, 95% CI 0.860, 0.955). The calibration curve and decision curve analysis showed that the nomogram helped differentiate MPE from benign pleural effusion (BPE) in clinical practice.

CONCLUSION

This study presents a nomogram model incorporating CT scan features and radiomics features to help physicians differentiate MPE from BPE.

Immunological characteristics of immunogenic cell death genes and malignant progression driving roles of TLR4 in anaplastic thyroid carcinoma

Anaplastic thyroid carcinoma (ATC) was a rare malignancy featured with the weak immunotherapeutic response. So far, disorders of immunogenic cell death genes (ICDGs) were identified as the driving factors in cancer progression, while their roles in ATC remained poorly clear. Datasets analysis identified that most ICDGs were high expressed in ATC, while DE-ICDGs were located in module c1_112, which was mainly enriched in Toll-like receptor signalings. Subsequently, the ICD score was established to classify ATC samples into the high and low ICD score groups, and function analysis indicated that high ICD score was associated with the immune characteristics. The high ICD score group had higher proportions of specific immune and stromal cells, as well as increased expression of immune checkpoints. Additionally, TLR4, ENTPD1, LY96, CASP1 and PDIA3 were identified as the dynamic signature in the malignant progression of ATC. Notably, TLR4 was significantly upregulated in ATC tissues, associated with poor prognosis. Silence of TLR4 inhibited the proliferation, metastasis and clone formation of ATC cells. Eventually, silence of TLR4 synergistically enhanced paclitaxel-induced proliferation inhibition, apoptosis, CALR exposure and release of ATP. Our findings highlighted that the aberrant expression of TLR4 drove the malignant progression of ATC, which contributed to our understanding of the roles of ICDGs in ATC.

Network medicine-based epistasis detection in complex diseases: ready for quantum computing

Most heritable diseases are polygenic. To comprehend the underlying genetic architecture, it is crucial to discover the clinically relevant epistatic interactions (EIs) between genomic single nucleotide polymorphisms (SNPs)1-3. Existing statistical computational methods for EI detection are mostly limited to pairs of SNPs due to the combinatorial explosion of higher-order EIs. With NeEDL (network-based epistasis detection via local search), we leverage network medicine to inform the selection of EIs that are an order of magnitude more statistically significant compared to existing tools and consist, on average, of five SNPs. We further show that this computationally demanding task can be substantially accelerated once quantum computing hardware becomes available. We apply NeEDL to eight different diseases and discover genes (affected by EIs of SNPs) that are partly known to affect the disease, additionally, these results are reproducible across independent cohorts. EIs for these eight diseases can be interactively explored in the Epistasis Disease Atlas (https://epistasis-disease-atlas.com). In summary, NeEDL is the first application that demonstrates the potential of seamlessly integrated quantum computing techniques to accelerate biomedical research. Our network medicine approach detects higher-order EIs with unprecedented statistical and biological evidence, yielding unique insights into polygenic diseases and providing a basis for the development of improved risk scores and combination therapies.

Improving power of genome-wide association studies via transforming ordinal phenotypes into continuous phenotypes

Introduction

Ordinal traits are important complex traits in crops, while genome-wide association study (GWAS) is a widely-used method in their gene mining. Presently, GWAS of continuous quantitative traits (C-GWAS) and single-locus association analysis method of ordinal traits are the main methods used for ordinal traits. However, the detection power of these two methods is low.

Methods

To address this issue, we proposed a new method, named MTOTC, in which hierarchical data of ordinal traits are transformed into continuous phenotypic data (CPData).

Results

Then, FASTmrMLM, one C-GWAS method, was used to conduct GWAS for CPData. The results from the simulation studies showed that, MTOTC+FASTmrMLM for ordinal traits was better than the classical methods when there were four and fewer hierarchical levels. In addition, when MTOTC was combined with FASTmrEMMA, mrMLM, ISIS EM-BLASSO, pLARmEB, and pKWmEB, relatively high power and low false positive rate in QTN detection were observed as well. Subsequently, MTOTC was applied to analyze the hierarchical data of soybean salt-alkali tolerance. It was revealed that more significant QTNs were detected when MTOTC was combined with any of the above six C-GWAs.

Discussion

Accordingly, the new method increases the choices of the GWAS methods for ordinal traits and helps to mine the genes for ordinal traits in resource populations.

A urine extracellular vesicle lncRNA classifier for high-grade prostate cancer and increased risk of progression: A multi-center study

To construct a urine extracellular vesicle long non-coding RNA (lncRNA) classifier that can detect high-grade prostate cancer (PCa) of grade group 2 or greater and estimate the risk of progression during active surveillance, we identify high-grade PCa-specific lncRNAs by combined analyses of cohorts from TAHSY, TCGA, and the GEO database. We develop and validate a 3-lncRNA diagnostic model (Clnc, being made of AC015987.1, CTD-2589M5.4, RP11-363E6.3) that can detect high-grade PCa. Clnc shows higher accuracy than prostate cancer antigen 3 (PCA3), multiparametric magnetic resonance imaging (mpMRI), and two risk calculators (Prostate Cancer Prevention Trial [PCPT]-RC 2.0 and European Randomized Study of Screening for Prostate Cancer [ERSPC]-RC) in the training cohort (n = 350), two independent cohorts (n = 232; n = 251), and TCGA cohort (n = 499). In the prospective active surveillance cohort (n = 182), Clnc at diagnosis remains a powerful independent predictor for overall active surveillance progression. Thus, Clnc is a potential biomarker for high-grade PCa and can also serve as a biomarker for improved selection of candidates for active surveillance.

New statistical selection method for pleiotropic variants associated with both quantitative and qualitative traits

BACKGROUND

Identification of pleiotropic variants associated with multiple phenotypic traits has received increasing attention in genetic association studies. Overlapping genetic associations from multiple traits help to detect weak genetic associations missed by single-trait analyses. Many statistical methods were developed to identify pleiotropic variants with most of them being limited to quantitative traits when pleiotropic effects on both quantitative and qualitative traits have been observed. This is a statistically challenging problem because there does not exist an appropriate multivariate distribution to model both quantitative and qualitative data together. Alternatively, meta-analysis methods can be applied, which basically integrate summary statistics of individual variants associated with either a quantitative or a qualitative trait without accounting for correlations among genetic variants.

RESULTS

We propose a new statistical selection method based on a unified selection score quantifying how a genetic variant, i.e., a pleiotropic variant associates with both quantitative and qualitative traits. In our extensive simulation studies where various types of pleiotropic effects on both quantitative and qualitative traits were considered, we demonstrated that the proposed method outperforms the existing meta-analysis methods in terms of true positive selection. We also applied the proposed method to a peanut dataset with 6 quantitative and 2 qualitative traits, and a cowpea dataset with 2 quantitative and 6 qualitative traits. We were able to detect some potentially pleiotropic variants missed by the existing methods in both analyses.

CONCLUSIONS

The proposed method is able to locate pleiotropic variants associated with both quantitative and qualitative traits. It has been implemented into an R package 'UNISS', which can be downloaded from http://github.com/statpng/uniss.

Associations of dietary, sociodemographic, and anthropometric factors with anemia among the Zhuang ethnic adults: a cross-sectional study in Guangxi Zhuang Autonomous Region, China

BACKGROUND

After decades of rapid economic development, anemia remains a significant public health challenge globally. This study aimed to estimate the associations of sociodemographic, dietary, and body composition factors with anemia among the Zhuang in Guangxi Zhuang Autonomous Region, China.

METHODS

Our study population from the baseline survey of the Guangxi ethnic minority Cohort Study of Chronic Diseases consisted of 13,465 adults (6,779 women and 6,686 men) aged 24-82 years. A validated interviewer-administered laptop-based questionnaire system was used to collect information on participants' sociodemographic, lifestyle, and dietary factors. Each participant underwent a physical examination, and hematological indices were measured. Least absolute shrinkage and selection operator (LASSO) regression was used to select the variables, and logistic regression was applied to estimate the associations of independent risk factors with anemia.

RESULTS

The overall prevalences of anemia in men and women were 9.63% (95% CI: 8.94-10.36%) and 18.33% (95% CI: 17.42─19.28%), respectively. LASSO and logistic regression analyses showed that age was positively associated with anemia for both women and men. For diet in women, red meat consumption for 5-7 days/week (OR = 0.79, 95% CI: 0.65-0.98, p = 0.0290) and corn/sweet potato consumption for 5-7 days/week (OR = 0.73, 95% CI: 0.55-0.96, p = 0.0281) were negatively associated with anemia. For men, fruit consumption for 5-7 days/week (OR = 0.75, 95% CI: 0.60-0.94, p = 0.0130) and corn/sweet potato consumption for 5-7 days/week (OR = 0.66, 95% CI: 0.46-0.91, p = 0.0136) were negatively correlated with anemia. Compared with a normal body water percentage (55-65%), a body water percentage below normal (< 55%) was negatively related to anemia (OR = 0.68, 95% CI: 0.53-0.86, p = 0.0014). Conversely, a body water percentage above normal (> 65%) was positively correlated with anemia in men (OR = 1.73, 95% CI: 1.38-2.17, p < 0.0001).

CONCLUSIONS

Anemia remains a moderate public health problem for premenopausal women and the elderly population in the Guangxi Zhuang minority region. The prevention of anemia at the population level requires multifaceted intervention measures according to sex and age, with a focus on dietary factors and the control of body composition.

oFVSD: a Python package of optimized forward variable selection decoder for high-dimensional neuroimaging data

The complexity and high dimensionality of neuroimaging data pose problems for decoding information with machine learning (ML) models because the number of features is often much larger than the number of observations. Feature selection is one of the crucial steps for determining meaningful target features in decoding; however, optimizing the feature selection from such high-dimensional neuroimaging data has been challenging using conventional ML models. Here, we introduce an efficient and high-performance decoding package incorporating a forward variable selection (FVS) algorithm and hyper-parameter optimization that automatically identifies the best feature pairs for both classification and regression models, where a total of 18 ML models are implemented by default. First, the FVS algorithm evaluates the goodness-of-fit across different models using the k-fold cross-validation step that identifies the best subset of features based on a predefined criterion for each model. Next, the hyperparameters of each ML model are optimized at each forward iteration. Final outputs highlight an optimized number of selected features (brain regions of interest) for each model with its accuracy. Furthermore, the toolbox can be executed in a parallel environment for efficient computation on a typical personal computer. With the optimized forward variable selection decoder (oFVSD) pipeline, we verified the effectiveness of decoding sex classification and age range regression on 1,113 structural magnetic resonance imaging (MRI) datasets. Compared to ML models without the FVS algorithm and with the Boruta algorithm as a variable selection counterpart, we demonstrate that the oFVSD significantly outperformed across all of the ML models over the counterpart models without FVS (approximately 0.20 increase in correlation coefficient, r, with regression models and 8% increase in classification models on average) and with Boruta variable selection algorithm (approximately 0.07 improvement in regression and 4% in classification models). Furthermore, we confirmed the use of parallel computation considerably reduced the computational burden for the high-dimensional MRI data. Altogether, the oFVSD toolbox efficiently and effectively improves the performance of both classification and regression ML models, providing a use case example on MRI datasets. With its flexibility, oFVSD has the potential for many other modalities in neuroimaging. This open-source and freely available Python package makes it a valuable toolbox for research communities seeking improved decoding accuracy.

Bayesian bi-level variable selection for genome-wide survival study

Mild cognitive impairment (MCI) is a clinical syndrome characterized by the onset and evolution of cognitive impairments, often considered a transitional stage to Alzheimer's disease (AD). The genetic traits of MCI patients who experience a rapid progression to AD can enhance early diagnosis capabilities and facilitate drug discovery for AD. While a genome-wide association study (GWAS) is a standard tool for identifying single nucleotide polymorphisms (SNPs) related to a disease, it fails to detect SNPs with small effect sizes due to stringent control for multiple testing. Additionally, the method does not consider the group structures of SNPs, such as genes or linkage disequilibrium blocks, which can provide valuable insights into the genetic architecture. To address the limitations, we propose a Bayesian bi-level variable selection method that detects SNPs associated with time of conversion from MCI to AD. Our approach integrates group inclusion indicators into an accelerated failure time model to identify important SNP groups. Additionally, we employ data augmentation techniques to impute censored time values using a predictive posterior. We adapt Dirichlet-Laplace shrinkage priors to incorporate the group structure for SNP-level variable selection. In the simulation study, our method outperformed other competing methods regarding variable selection. The analysis of Alzheimer's Disease Neuroimaging Initiative (ADNI) data revealed several genes directly or indirectly related to AD, whereas a classical GWAS did not identify any significant SNPs.

Weighted multiple testing procedures in genome-wide association studies

Multiple testing procedures controlling the false discovery rate (FDR) are increasingly used in the context of genome wide association studies (GWAS), and weighted multiple testing procedures that incorporate covariate information are efficient to improve the power to detect associations. In this work, we evaluate some recent weighted multiple testing procedures in the specific context of GWAS through a simulation study. We also present a new efficient procedure called wBHa that prioritizes the detection of genetic variants with low minor allele frequencies while maximizing the overall detection power. The results indicate good performance of our procedure compared to other weighted multiple testing procedures. In particular, in all simulated settings, wBHa tends to outperform other procedures in detecting rare variants while maintaining good overall power. The use of the different procedures is illustrated with a real dataset.

Learning Peptide Properties with Positive Examples Only

Deep learning can create accurate predictive models by exploiting existing large-scale experimental data, and guide the design of molecules. However, a major barrier is the requirement of both positive and negative examples in the classical supervised learning frameworks. Notably, most peptide databases come with missing information and low number of observations on negative examples, as such sequences are hard to obtain using high-throughput screening methods. To address this challenge, we solely exploit the limited known positive examples in a semi-supervised setting, and discover peptide sequences that are likely to map to certain antimicrobial properties via positive-unlabeled learning (PU). In particular, we use the two learning strategies of adapting base classifier and reliable negative identification to build deep learning models for inferring solubility, hemolysis, binding against SHP-2, and non-fouling activity of peptides, given their sequence. We evaluate the predictive performance of our PU learning method and show that by only using the positive data, it can achieve competitive performance when compared with the classical positive-negative (PN) classification approach, where there is access to both positive and negative examples.

Combining phenotypic and genomic data to improve prediction of binary traits

Plant breeders want to develop cultivars that outperform existing genotypes. Some characteristics (here 'main traits') of these cultivars are categorical and difficult to measure directly. It is important to predict the main trait of newly developed genotypes accurately. In addition to marker data, breeding programs often have information on secondary traits (or 'phenotypes') that are easy to measure. Our goal is to improve prediction of main traits with interpretable relations by combining the two data types using variable selection techniques. However, the genomic characteristics can overwhelm the set of secondary traits, so a standard technique may fail to select any phenotypic variables. We develop a new statistical technique that ensures appropriate representation from both the secondary traits and the genotypic variables for optimal prediction. When two data types (markers and secondary traits) are available, we achieve improved prediction of a binary trait by two steps that are designed to ensure that a significant intrinsic effect of a phenotype is incorporated in the relation before accounting for extra effects of genotypes. First, we sparsely regress the secondary traits on the markers and replace the secondary traits by their residuals to obtain the effects of phenotypic variables as adjusted by the genotypic variables. Then, we develop a sparse logistic classifier using the markers and residuals so that the adjusted phenotypes may be selected first to avoid being overwhelmed by the genotypic variables due to their numerical advantage. This classifier uses forward selection aided by a penalty term and can be computed effectively by a technique called the one-pass method. It compares favorably with other classifiers on simulated and real data.

Multivariate genome-wide association analysis by iterative hard thresholding

MOTIVATION

In a genome-wide association study, analyzing multiple correlated traits simultaneously is potentially superior to analyzing the traits one by one. Standard methods for multivariate genome-wide association study operate marker-by-marker and are computationally intensive.

RESULTS

We present a sparsity constrained regression algorithm for multivariate genome-wide association study based on iterative hard thresholding and implement it in a convenient Julia package MendelIHT.jl. In simulation studies with up to 100 quantitative traits, iterative hard thresholding exhibits similar true positive rates, smaller false positive rates, and faster execution times than GEMMA's linear mixed models and mv-PLINK's canonical correlation analysis. On UK Biobank data with 470 228 variants, MendelIHT completed a three-trait joint analysis (n=185 656) in 20 h and an 18-trait joint analysis (n=104 264) in 53 h with an 80 GB memory footprint. In short, MendelIHT enables geneticists to fit a single regression model that simultaneously considers the effect of all SNPs and dozens of traits.

AVAILABILITY AND IMPLEMENTATION

Software, documentation, and scripts to reproduce our results are available from https://github.com/OpenMendel/MendelIHT.jl.

Integrating and optimizing genomic, weather, and secondary trait data for multiclass classification

Modern plant breeding programs collect several data types such as weather, images, and secondary or associated traits besides the main trait (e.g., grain yield). Genomic data is high-dimensional and often over-crowds smaller data types when naively combined to explain the response variable. There is a need to develop methods able to effectively combine different data types of differing sizes to improve predictions. Additionally, in the face of changing climate conditions, there is a need to develop methods able to effectively combine weather information with genotype data to predict the performance of lines better. In this work, we develop a novel three-stage classifier to predict multi-class traits by combining three data types-genomic, weather, and secondary trait. The method addressed various challenges in this problem, such as confounding, differing sizes of data types, and threshold optimization. The method was examined in different settings, including binary and multi-class responses, various penalization schemes, and class balances. Then, our method was compared to standard machine learning methods such as random forests and support vector machines using various classification accuracy metrics and using model size to evaluate the sparsity of the model. The results showed that our method performed similarly to or better than machine learning methods across various settings. More importantly, the classifiers obtained were highly sparse, allowing for a straightforward interpretation of relationships between the response and the selected predictors.

An efficient model-free approach to interaction screening for high dimensional data

An innovated model-free interaction screening procedure called the MCVIS is proposed for high dimensional data analysis. Specifically, we adopt the introduced MCV index for quantifying the importance of an interaction effect among predictors. Our proposed method is fully nonparametric and is capable of successfully selecting interactions even if the signal of parental main effects is weak. The MCVIS procedure has many distinctive features: (i) it can work with discrete, categorical and continuous covariates; (ii) it can deal with both categorical and continuous response, even handle the missing response; (iii) it is robust for heavy-tailed distributions, thus well accommodates heterogeneity typically caused by high dimensionality; (iv) it enjoys the sure screening and ranking consistency properties, therefore achieves dimension reduction without information loss. In another respect, computational feasibility is a top concern in high dimensional data analysis, by transforming our MCV into several variants, the MCVIS procedure is simple and fast to implement. Extensive numerical experiments and comparisons confirm the effectiveness and wide applicability of our MCVIS procedure. We further illustrate the proposed methodology by empirical study of two real datasets. Supplementary materials are available online.

mRNA transport, translation, and decay in adult mammalian central nervous system axons

Localized mRNA translation regulates synapse function and axon maintenance, but how compartment-specific mRNA repertoires are regulated is largely unknown. We developed an axonal transcriptome capture method that allows deep sequencing of metabolically labeled mRNAs from retinal ganglion cell axon terminals in mouse. Comparing axonal-to-somal transcriptomes and axonal translatome-to-transcriptome enables genome-wide visualization of mRNA transport and translation and unveils potential regulators tuned to each process. FMRP and TDP-43 stand out as key regulators of transport, and experiments in Fmr1 knockout mice validate FMRP's role in the axonal transportation of synapse-related mRNAs. Pulse-and-chase experiments enable genome-wide assessment of mRNA stability in axons and reveal a strong coupling between mRNA translation and decay. Measuring the absolute mRNA abundance per axon terminal shows that the adult axonal transcriptome is stably maintained by persistent transport. Our datasets provide a rich resource for unique insights into RNA-based mechanisms in maintaining presynaptic structure and function in vivo.

Testing generalized linear models with high-dimensional nuisance parameter

Generalized linear models often have a high-dimensional nuisance parameters, as seen in applications such as testing gene-environment interactions or gene-gene interactions. In these scenarios, it is essential to test the significance of a high-dimensional sub-vector of the model's coefficients. Although some existing methods can tackle this problem, they often rely on the bootstrap to approximate the asymptotic distribution of the test statistic, and thus are computationally expensive. Here, we propose a computationally efficient test with a closed-form limiting distribution, which allows the parameter being tested to be either sparse or dense. We show that under certain regularity conditions, the type I error of the proposed method is asymptotically correct, and we establish its power under high-dimensional alternatives. Extensive simulations demonstrate the good performance of the proposed test and its robustness when certain sparsity assumptions are violated. We also apply the proposed method to Chinese famine sample data in order to show its performance when testing the significance of gene-environment interactions.

Multicenter clinical radiomics-integrated model based on [18F]FDG PET and multi-modal MRI predict ATRX mutation status in IDH-mutant lower-grade gliomas

OBJECTIVES

To develop a clinical radiomics-integrated model based on ¹⁸ F-fluorodeoxyglucose positron emission tomography ([¹⁸F]FDG PET) and multi-modal MRI for predicting alpha thalassemia/mental retardation X-linked (ATRX) mutation status of IDH-mutant lower-grade gliomas (LGGs).

METHODS

One hundred and two patients (47 ATRX mutant-type, 55 ATRX wild-type) diagnosed with IDH-mutant LGGs (CNS WHO grades 1 and 2) were retrospectively enrolled. A total of 5540 radiomics features were extracted from structural MR (sMR) images (contrast-enhanced T1-weighted imaging, CE-T1WI; T2-weighted imaging, and T2WI), functional MR (fMR) images (apparent diffusion coefficient, ADC; cerebral blood volume, CBV), and metabolic PET images ([¹⁸F]FDG PET). The random forest algorithm was used to establish a clinical radiomics-integrated model, integrating the optimal multi-modal radiomics model with three clinical parameters. The predictive effectiveness of the models was evaluated by receiver operating characteristic (ROC) and decision curve analysis (DCA).

RESULTS

The optimal multi-modal model incorporated sMR (CE-T1WI), fMR (ADC), and metabolic ([¹⁸F]FDG) images ([¹⁸F]FDG PET+ADC+ CE-T1WI) with the area under curves (AUCs) in the training and test groups of 0.971 and 0.962, respectively. The clinical radiomics-integrated model, incorporating [¹⁸F]FDG PET+ADC+CE-T1WI, three clinical parameters (KPS, SFSD, and ATGR), showed the best predictive effectiveness in the training and test groups (0.987 and 0.975, respectively).

CONCLUSIONS

The clinical radiomics-integrated model with metabolic, structural, and functional information based on [¹⁸F]FDG PET and multi-modal MRI achieved promising performance for predicting the ATRX mutation status of IDH-mutant LGGs.

KEY POINTS

• The clinical radiomics-integrated model based on [¹⁸F]FDG PET and multi-modal MRI achieved promising performance for predicting ATRX mutation status in LGGs. • The study investigated the value of multicenter clinical radiomics-integrated model based on [¹⁸F]FDG PET and multi-modal MRI in LGGs regarding ATRX mutation status prediction. • The integrated model provided structural, functional, and metabolic information simultaneously and demonstrated with satisfactory calibration and discrimination in the training and test groups (0.987 and 0.975, respectively).

MP-LASSO chart: a multi-level polar chart for visualizing group LASSO analysis of genomic data

Penalized regression has been widely used in genome-wide association studies for jointanalyses to find genetic associations. Among penalized regression models, the least absolute shrinkage and selection operator (Lasso) method effectively removes some coefficientsfrom the model by shrinking them to zero. To handle group structures, such as genes andpathways, several modified Lasso penalties have been proposed, including group Lasso andsparse group Lasso. Group Lasso ensures sparsity at the level of pre-defined groups, eliminating unimportant groups. Sparse group Lasso performs group selection as in group Lasso,but also performs individual selection as in Lasso. While these sparse methods are useful inhigh-dimensional genetic studies, interpreting the results with many groups and coefficients is not straightforward. Lasso's results are often expressed as trace plots of regressioncoefficients. However, few studies have explored the systematic visualization of group information. In this study, we propose a multi-level polar Lasso (MP-Lasso) chart, which caneffectively represent the results from group Lasso and sparse group Lasso analyses. An Rpackage to draw MP-Lasso charts was developed. Through a real-world genetic data application, we demonstrated that our MP-Lasso chart package effectively visualizes the resultsof Lasso, group Lasso, and sparse group Lasso.

StackCirRNAPred: computational classification of long circRNA from other lncRNA based on stacking strategy

BACKGROUND

CircRNAs are essential for the regulation of post-transcriptional gene expression, including as miRNA sponges, and play an important role in disease development. Some computational tools have been proposed recently to predict circRNA, since only one classifier is used, there is still much that can be done to improve the performance.

RESULTS

StackCirRNAPred was proposed, the computational classification of long circRNA from other lncRNA based on stacking strategy. In order to cope with the potential problem that a single feature might not be able to distinguish circRNA well from other lncRNA, we first extracted features from different sources, including nucleic acid composition, sequence spatial features and physicochemical properties, Alu and tandem repeats. We innovatively apply the stacking strategy to integrate the more advantageous classifiers of RF, LightGBM, XGBoost. This allows the model to incorporate these features more flexibly. StackCirRNAPred was found to be significantly better than other tools, with precision, accuracy, F1, recall and MCC of 0.843, 0.833, 0.831, 0.819 and 0.666 respectively. We tested it directly on the mouse dataset. StackCirRNAPred was still significantly better than other methods, with precision, accuracy, F1, recall and MCC of 0.837, 0.839, 0.839, 0.841, 0.677.

CONCLUSIONS

We proposed StackCirRNAPred based on stacking strategy to distinguish long circRNAs from other lncRNAs. With the test results demonstrating the validity and robustness of StackCirRNAPred, we hope StackCirRNAPred will complement existing circRNA prediction methods and is helpful in down-stream research.

A Poisson reduced-rank regression model for association mapping in sequencing data

BACKGROUND

Single-cell RNA-sequencing (scRNA-seq) technologies allow for the study of gene expression in individual cells. Often, it is of interest to understand how transcriptional activity is associated with cell-specific covariates, such as cell type, genotype, or measures of cell health. Traditional approaches for this type of association mapping assume independence between the outcome variables (or genes), and perform a separate regression for each. However, these methods are computationally costly and ignore the substantial correlation structure of gene expression. Furthermore, count-based scRNA-seq data pose challenges for traditional models based on Gaussian assumptions.

RESULTS

We aim to resolve these issues by developing a reduced-rank regression model that identifies low-dimensional linear associations between a large number of cell-specific covariates and high-dimensional gene expression readouts. Our probabilistic model uses a Poisson likelihood in order to account for the unique structure of scRNA-seq counts. We demonstrate the performance of our model using simulations, and we apply our model to a scRNA-seq dataset, a spatial gene expression dataset, and a bulk RNA-seq dataset to show its behavior in three distinct analyses.

CONCLUSION

We show that our statistical modeling approach, which is based on reduced-rank regression, captures associations between gene expression and cell- and sample-specific covariates by leveraging low-dimensional representations of transcriptional states.

CRISPR/Cas9 technology as an innovative approach to enhancing the phytoremediation: Concepts and implications

Phytoremediation is currently an active field of research focusing chiefly on identifying and characterizing novel and high chelation action super-accumulators. In the last few years, molecular tools have been widely exploited to understand better metal absorption, translocation, cation, and tolerance mechanisms in plants. Recently more advanced CRISPR-Cas9 genome engineering technology is also employed to enhance detoxification efficiency. Further, advances in molecular science will trigger the understanding of adaptive phytoremediation ability plant production in current global warming conditions. The enhanced abilities of nucleases for genome modification can improve plant repair capabilities by modifying the genome, thereby achieving a sustainable ecosystem. The purpose of this manuscript focuses on biotechnology's fundamental principles and application to promote climate-resistant metal plants, especially the CRISPR-Cas9 genome editing system for enhancing the phytoremediation of harmful contamination and pollutants.

A Novel Multitasking Ant Colony Optimization Method for Detecting Multiorder SNP Interactions

MOTIVATION

Linear or nonlinear interactions of multiple single-nucleotide polymorphisms (SNPs) play an important role in understanding the genetic basis of complex human diseases. However, combinatorial analytics in high-dimensional space makes it extremely challenging to detect multiorder SNP interactions. Most classic approaches can only perform one task (for detecting k-order SNP interactions) in each run. Since prior knowledge of a complex disease is usually not available, it is difficult to determine the value of k for detecting k-order SNP interactions.

METHODS

A novel multitasking ant colony optimization algorithm (named MTACO-DMSI) is proposed to detect multiorder SNP interactions, and it is divided into two stages: searching and testing. In the searching stage, multiple multiorder SNP interaction detection tasks (from 2nd-order to kth-order) are executed in parallel, and two subpopulations that separately adopt the Bayesian network-based K2-score and Jensen-Shannon divergence (JS-score) as evaluation criteria are generated for each task to improve the global search capability and the discrimination ability for various disease models. In the testing stage, the G test statistical test is adopted to further verify the authenticity of candidate solutions to reduce the error rate.

RESULT

Three multiorder simulated disease models with different interaction effects and three real age-related macular degeneration (AMD), rheumatoid arthritis (RA) and type 1 diabetes (T1D) datasets were used to investigate the performance of the proposed MTACO-DMSI. The experimental results show that the MTACO-DMSI has a faster search speed and higher discriminatory power for diverse simulation disease models than traditional single-task algorithms. The results on real AMD data and RA and T1D datasets indicate that MTACO-DMSI has the ability to detect multiorder SNP interactions at a genome-wide scale. Availability and implementation: https://github.com/shouhengtuo/MTACO-DMSI/.

A penalized linear mixed model with generalized method of moments estimators for complex phenotype prediction

MOTIVATION

Linear mixed models (LMMs) have long been the method of choice for risk prediction analysis on high-dimensional data. However, it remains computationally challenging to simultaneously model a large amount of variants that can be noise or have predictive effects of complex forms.

RESULTS

In this work, we have developed a penalized LMM with generalized method of moments (pLMMGMM) estimators for prediction analysis. pLMMGMM is built within the LMM framework, where random effects are used to model the joint predictive effects from all variants within a region. Different from existing methods that focus on linear relationships and use empirical criteria for variable screening, pLMMGMM can efficiently detect regions that harbor genetic variants with both linear and non-linear predictive effects. In addition, unlike existing LMMs that can only handle a very limited number of random effects, pLMMGMM is much less computationally demanding. It can jointly consider a large number of regions and accurately detect those that are predictive. Through theoretical investigations, we have shown that our method has the selection consistency and asymptotic normality. Through extensive simulations and the analysis of PET-imaging outcomes, we have demonstrated that pLMMGMM outperformed existing models and it can accurately detect regions that harbor risk factors with various forms of predictive effects.

AVAILABILITY AND IMPLEMENTATION

The R-package is available at https://github.com/XiaQiong/GMMLasso.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Development and Validation of a Model Using Radiomics Features from an Apparent Diffusion Coefficient Map to Diagnose Local Tumor Recurrence in Patients Treated for Head and Neck Squamous Cell Carcinoma

OBJECTIVE

To develop and validate a model using radiomics features from apparent diffusion coefficient (ADC) map to diagnose local tumor recurrence in head and neck squamous cell carcinoma (HNSCC).

MATERIALS AND METHODS

This retrospective study included 285 patients (mean age ± standard deviation, 62 ± 12 years; 220 male, 77.2%), including 215 for training (n = 161) and internal validation (n = 54) and 70 others for external validation, with newly developed contrast-enhancing lesions at the primary cancer site on the surveillance MRI following definitive treatment of HNSCC between January 2014 and October 2019. Of the 215 and 70 patients, 127 and 34, respectively, had local tumor recurrence. Radiomics models using radiomics scores were created separately for T2-weighted imaging (T2WI), contrast-enhanced T1-weighted imaging (CE-T1WI), and ADC maps using non-zero coefficients from the least absolute shrinkage and selection operator in the training set. Receiver operating characteristic (ROC) analysis was used to evaluate the diagnostic performance of each radiomics score and known clinical parameter (age, sex, and clinical stage) in the internal and external validation sets.

RESULTS

Five radiomics features from T2WI, six from CE-T1WI, and nine from ADC maps were selected and used to develop the respective radiomics models. The area under ROC curve (AUROC) of ADC radiomics score was 0.76 (95% confidence interval [CI], 0.62-0.89) and 0.77 (95% CI, 0.65-0.88) in the internal and external validation sets, respectively. These were significantly higher than the AUROC values of T2WI (0.53 [95% CI, 0.40-0.67], p = 0.006), CE-T1WI (0.53 [95% CI, 0.40-0.67], p = 0.012), and clinical parameters (0.53 [95% CI, 0.39-0.67], p = 0.021) in the external validation set.

CONCLUSION

The radiomics model using ADC maps exhibited higher diagnostic performance than those of the radiomics models using T2WI or CE-T1WI and clinical parameters in the diagnosis of local tumor recurrence in HNSCC following definitive treatment.

The Construction of ITP Diagnostic Modeling Based on the Expressions of Hub Genes Associated with M1 Polarization of Macrophages

Purpose

Primary immune thrombocytopenia (ITP) is an immune disease with a diagnosis of exclusion, since no validated biomarkers have been identified. In this study, we explored biomarkers associated with the development of ITP from an immune perspective to inform the clinical diagnosis.

Patients and Methods

Differentially expressed genes (DEGs) between normal and ITP samples were analyzed using limma package. Random forest algorithm and LASSO regression were further used to screen for DEGs associated with ITP. The expression of these hub genes was validated by PCR. The relationship between DEGs and immunity was explored by enrichment analysis. Immune cell infiltration in ITP was analyzed by CIBERSORT and ssGSEA, and the relationship between DEGs and infiltrating immune cells was analyzed by Spearman’s rank correlation analysis. Finally, a diagnostic model related to DEGs was constructed by the neural network, and its efficiency was detected by the ROC curve.

Results

After screening the GEO database and validation by PCR analysis, The expression of CTH and TAF8 were higher and while OSBP2 expression was lower in ITP patients compared to normal subjects (P<0.05). GO enrichment analysis showed that these DEGs were associated with inflammatory immune-related diseases, and KEGG analysis showed that they mainly regulated signaling pathways such as JAK-STAT. CIBERSORT and ssGSEA analyses showed that these DEGs were mainly associated with macrophage M1 polarization. The expression of CTH and TAF8 were positively correlated with M1 expression, while OSBP2 was negatively correlated with M1 expression. The ROC curve showed high accuracy of the neural network model [AUC= 0.939, 95% CI (0.8–1)].

Conclusion

Our results suggest that CTH, TAF8, and OSBP2 can be used as effective diagnostic biomarkers of ITP.

Identification and Validation of an Inflammatory Response-Related Polygenic Risk Score as a Prognostic Marker in Hepatocellular Carcinoma

Aims

We hypothesized that the expression patterns of inflammatory response-related genes may be a potential tool for hepatocellular carcinoma (HCC) risk scoring.

Background

Inflammatory response plays a pivotal role in the pathogenesis of HCC.

Objective

To establish and validate a hallmark inflammatory response gene-based polygenic risk score as a prognostic tool in HCC.

Methods

We screened differentially expressed inflammatory response genes and established an inflammatory response-related polygenic risk score (IRPRS) in an HCC-related dataset. Patients with HCC were categorized into high- and low-risk groups according to the median IRPRS, and the overall survival between the two groups was compared. The IRPRS was validated in an independent external dataset. Tumor-infiltrating lymphocytes (TILs) in high- and low-risk groups were compared, and gene set enrichment analysis was performed to characterize high-risk HCC identified using this IRPRS.

Results

Four differentially expressed hallmark inflammatory response genes (CD14, AQP9, SERPINE1, and ITGA5) were identified to construct the IRPRS. Patients in the high-risk group had significantly shorter overall survival than those in the low-risk group in both the training set and the test set. Furthermore, the IRPRS remained an independent prognostic factor compared to the routine clinicopathological characteristics. Many cancer-related hallmark gene sets and TILs were significantly enriched in the high-risk group.

Conclusions

We established and validated a four-hallmark inflammatory response gene-based polygenic risk score, which could successfully divide patients with HCC into high-risk and low-risk groups. These two risk groups of HCC possess significantly distinct prognostic and biological characteristics.

Classification of COVID19 Patients Using Robust Logistic Regression

Coronavirus disease 2019 (COVID19) has triggered a global pandemic affecting millions of people. Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) causing the COVID-19 disease is hypothesized to gain entry into humans via the airway epithelium, where it initiates a host response. The expression levels of genes at the upper airway that interact with the SARS-CoV-2 could be a telltale sign of virus infection. However, gene expression data have been flagged as suspicious of containing different contamination errors via techniques for extracting such information, and clinical diagnosis may contain labelling errors due to the specificity and sensitivity of diagnostic tests. We propose to fit the regularized logistic regression model as a classifier for COVID-19 diagnosis, which simultaneously identifies genes related to the disease and predicts the COVID-19 cases based on the expression values of the selected genes. We apply a robust estimating methods based on the density power divergence to obtain stable results ignoring the effects of contamination or labelling errors in the data and compare its performance with respect to the classical maximum likelihood estimator with different penalties, including the LASSO and the general adaptive LASSO penalties.

Survival Analysis with High-Dimensional Omics Data Using a Threshold Gradient Descent Regularization-Based Neural Network Approach

Analysis of data with a censored survival response and high-dimensional omics measurements is now common. Most of the existing analyses are based on specific (semi)parametric models, in particular the Cox model. Such analyses may be limited by not having sufficient flexibility, for example, in accommodating nonlinearity. For categorical and continuous responses, neural networks (NNs) have provided a highly competitive alternative. Comparatively, NNs for censored survival data remain limited. Omics measurements are usually high-dimensional, and only a small subset is expected to be survival-associated. As such, regularized estimation and selection are needed. In the existing NN studies, this is usually achieved via penalization. In this article, we propose adopting the threshold gradient descent regularization (TGDR) technique, which has competitive performance (for example, when compared to penalization) and unique advantages in regression analysis, but has not been adopted with NNs. The TGDR-based NN has a highly sensible formulation and an architecture different from the unregularized and penalization-based ones. Simulations show its satisfactory performance. Its practical effectiveness is further established via the analysis of two cancer omics datasets. Overall, this study can provide a practical and useful new way in the NN paradigm for survival analysis with high-dimensional omics measurements.

Pancancer analysis of a potential gene mutation model in the prediction of immunotherapy outcomes

Background: Immune checkpoint blockade (ICB) represents a promising treatment for cancer, but predictive biomarkers are needed. We aimed to develop a cost-effective signature to predict immunotherapy benefits across cancers.

Methods: We proposed a study framework to construct the signature. Specifically, we built a multivariate Cox proportional hazards regression model with LASSO using 80% of an ICB-treated cohort (n = 1661) from MSKCC. The desired signature named SIGP was the risk score of the model and was validated in the remaining 20% of patients and an external ICB-treated cohort (n = 249) from DFCI.

Results: SIGP was based on 18 candidate genes (NOTCH3, CREBBP, RNF43, PTPRD, FAM46C, SETD2, PTPRT, TERT, TET1, ROS1, NTRK3, PAK7, BRAF, LATS1, IL7R, VHL, TP53, and STK11), and we classified patients into SIGP high (SIGP-H), SIGP low (SIGP-L) and SIGP wild type (SIGP-WT) groups according to the SIGP score. A multicohort validation demonstrated that patients in SIGP-L had significantly longer overall survival (OS) in the context of ICB therapy than those in SIGP-WT and SIGP-H (44.00 months versus 13.00 months and 14.00 months, p &lt; 0.001 in the test set). The survival of patients grouped by SIGP in non-ICB-treated cohorts was different, and SIGP-WT performed better than the other groups. In addition, SIGP-L + TMB-L (approximately 15% of patients) had similar survivals to TMB-H, and patients with both SIGP-L and TMB-H had better survival. Further analysis on tumor-infiltrating lymphocytes demonstrated that the SIGP-L group had significantly increased abundances of CD8+ T cells.

Conclusion: Our proposed model of the SIGP signature based on 18-gene mutations has good predictive value for the clinical benefit of ICB in pancancer patients. Additional patients without TMB-H were identified by SIGP as potential candidates for ICB, and the combination of both signatures showed better performance than the single signature.

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 penalized linear mixed model with generalized method of moments for prediction analysis on high-dimensional multi-omics data

With the advances in high-throughput biotechnologies, high-dimensional multi-layer omics data become increasingly available. They can provide both confirmatory and complementary information to disease risk and thus have offered unprecedented opportunities for risk prediction studies. However, the high-dimensionality and complex inter/intra-relationships among multi-omics data have brought tremendous analytical challenges. Here we present a computationally efficient penalized linear mixed model with generalized method of moments estimator (MpLMMGMM) for the prediction analysis on multi-omics data. Our method extends the widely used linear mixed model proposed for genomic risk predictions to model multi-omics data, where kernel functions are used to capture various types of predictive effects from different layers of omics data and penalty terms are introduced to reduce the impact of noise. Compared with existing penalized linear mixed models, the proposed method adopts the generalized method of moments estimator and it is much more computationally efficient. Through extensive simulation studies and the analysis of positron emission tomography imaging outcomes, we have demonstrated that MpLMMGMM can simultaneously consider a large number of variables and efficiently select those that are predictive from the corresponding omics layers. It can capture both linear and nonlinear predictive effects and achieves better prediction performance than competing methods.

Accurate Evaluation of Feature Contributions for Sentinel Lymph Node Status Classification in Breast Cancer

The current guidelines recommend the sentinel lymph node biopsy to evaluate the lymph node involvement for breast cancer patients with clinically negative lymph nodes on clinical or radiological examination. Machine learning (ML) models have significantly improved the prediction of lymph nodes status based on clinical features, thus avoiding expensive, time-consuming and invasive procedures. However, the classification of sentinel lymph node status represents a typical example of an unbalanced classification problem. In this work, we developed a ML framework to explore the effects of unbalanced populations on the performance and stability of feature ranking for sentinel lymph node status classification in breast cancer. Our results indicate state-of-the-art AUC (Area under the Receiver Operating Characteristic curve) values on a hold-out set (67%) while providing particularly stable features related to tumor size, histological subtype and estrogen receptor expression, which should therefore be considered as potential biomarkers.

Large-Scale Survey Data Analysis with Penalized Regression: A Monte Carlo Simulation on Missing Categorical Predictors

With the advent of the big data era, machine learning methods have evolved and proliferated. This study focused on penalized regression, a procedure that builds interpretive prediction models among machine learning methods. In particular, penalized regression coupled with large-scale data can explore hundreds or thousands of variables in one statistical model without convergence problems and identify yet uninvestigated important predictors. As one of the first Monte Carlo simulation studies to investigate predictive modeling with missing categorical predictors in the context of social science research, this study endeavored to emulate real social science large-scale data. Likert-scaled variables were simulated as well as multiple-category and count variables. Due to the inclusion of the categorical predictors in modeling, penalized regression methods that consider the grouping effect were employed such as group Mnet. We also examined the applicability of the simulation conditions with a real large-scale dataset that the simulation study referenced. Particularly, the study presented selection counts of variables after multiple iterations of modeling in order to consider the bias resulting from data-splitting in model validation. Selection counts turned out to be a necessary tool when variable selection is of research interest. Efforts to utilize large-scale data to the fullest appear to offer a valid approach to mitigate the effect of nonignorable missingness. Overall, penalized regression which assumes linearity is a viable method to analyze social science large-scale survey data.

Driver gene detection through Bayesian network integration of mutation and expression profiles

MOTIVATION

The identification of mutated driver genes and the corresponding pathways is one of the primary goals in understanding tumorigenesis at the patient level. Integration of multi-dimensional genomic data from existing repositories, e.g., The Cancer Genome Atlas (TCGA), offers an effective way to tackle this issue. In this study, we aimed to leverage the complementary genomic information of individuals and create an integrative framework to identify cancer-related driver genes. Specifically, based on pinpointed differentially expressed genes, variants in somatic mutations and a gene interaction network, we proposed an unsupervised Bayesian network integration (BNI) method to detect driver genes and estimate the disease propagation at the patient and/or cohort levels. This new method first captures inherent structural information to construct a functional gene mutation network and then extracts the driver genes and their controlled downstream modules using the minimum cover subset method.

RESULTS

Using other credible sources (e.g. Cancer Gene Census and Network of Cancer Genes), we validated the driver genes predicted by the BNI method in three TCGA pan-cancer cohorts. The proposed method provides an effective approach to address tumor heterogeneity faced by personalized medicine. The pinpointed drivers warrant further wet laboratory validation.

AVAILABILITY AND IMPLEMENTATION

The supplementary tables and source code can be obtained from https://xavieruniversityoflouisiana.sharefile.com/d-se6df2c8d0ebe4800a3030311efddafe5.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.