Novel deep learning-based solution for identification of prognostic subgroups in liver cancer (Hepatocellular carcinoma)

Background Liver cancer (Hepatocellular carcinoma; HCC) prevalence is increasing and with poor clinical outcome expected it means greater understanding of HCC aetiology is urgently required. This study explored a deep learning solution to detect biologically important features that distinguish prognostic subgroups. A novel architecture of an Artificial Neural Network (ANN) trained with a customised objective function (LRSC) was developed. The ANN should discover new data representations, to detect patient subgroups that are biologically homogenous (clustering loss) and similar in survival (survival loss) while removing noise from the data (reconstruction loss). The model was applied to TCGA-HCC multi-omics data and benchmarked against baseline models that only use a reconstruction objective function (BCE, MSE) for learning. With the baseline models, the new features are then filtered based on survival information and used for clustering patients. Different variants of the customised objective function, incorporating only reconstruction and clustering losses (LRC); and reconstruction and survival losses (LRS) were also evaluated. Robust features consistently detected were compared between models and validated in TCGA and LIRI-JP HCC cohorts. Results The combined loss (LRSC) discovered highly significant prognostic subgroups (P-value = 1.55E−77) with more accurate sample assignment (Silhouette scores: 0.59–0.7) compared to baseline models (0.18–0.3). All LRSC bottleneck features (N = 100) were significant for survival, compared to only 11–21 for baseline models. Prognostic subgroups were not explained by disease grade or risk factors. Instead LRSC identified robust features including 377 mRNAs, many of which were novel (61.27%) compared to those identified by the other losses. Some 75 mRNAs were prognostic in TCGA, while 29 were prognostic in LIRI-JP also. LRSC also identified 15 robust miRNAs including two novel (hsa-let-7g; hsa-mir-550a-1) and 328 methylation features with 71% being prognostic. Gene-enrichment and Functional Annotation Analysis identified seven pathways differentiating prognostic clusters. Conclusions Combining cluster and survival metrics with the reconstruction objective function facilitated superior prognostic subgroup identification. The hybrid model identified more homogeneous clusters that consequently were more biologically meaningful. The novel and prognostic robust features extracted provide additional information to improve our understanding of a complex disease to help reveal its aetiology. Moreover, the gene features identified may have clinical applications as therapeutic targets.

Subgroup identification in individual participant data meta-analysis using model-based recursive partitioning

Model-based recursive partitioning (MOB) can be used to identify subgroups with differing treatment effects. The detection rate of treatment-by-covariate interactions and the accuracy of identified subgroups using MOB depend strongly on the sample size. Using data from multiple randomized controlled clinical trials can overcome the problem of too small samples. However, naively pooling data from multiple trials may result in the identification of spurious subgroups as differences in study design, subject selection and other sources of between-trial heterogeneity are ignored. In order to account for between-trial heterogeneity in individual participant data (IPD) meta-analysis random-effect models are frequently used. Commonly, heterogeneity in the treatment effect is modelled using random effects whereas heterogeneity in the baseline risks is modelled by either fixed effects or random effects. In this article, we propose metaMOB, a procedure using the generalized mixed-effects model tree (GLMM tree) algorithm for subgroup identification in IPD meta-analysis. Although the application of metaMOB is potentially wider, e.g. randomized experiments with participants in social sciences or preclinical experiments in life sciences, we focus on randomized controlled clinical trials. In a simulation study, metaMOB outperformed GLMM trees assuming a random intercept only and model-based recursive partitioning (MOB), whose algorithm is the basis for GLMM trees, with respect to the false discovery rates, accuracy of identified subgroups and accuracy of estimated treatment effect. The most robust and therefore most promising method is metaMOB with fixed effects for modelling the between-trial heterogeneity in the baseline risks.

An interactive tool for identifying patient subgroups based on arbitrary characteristics for medical research

We present an analysis tool for subgroup identification in medical research based on feature analysis. Our use case is intracranial aneurysms. In the tool, an aneurysm-of-interest’s most similar aneurysms within a database are found. Similarity is defined via user-selected parameters, which can be entirely arbitrary. Different interactive outputs and visualizations include a heatmap view and a graph, which give an intuitive feedback to support researchers in the consideration of research questions, which in the present use case often relate to rupture risk analysis. The tool was evaluated with a pilot study and phantom database and received favorable results for its requirements of reliability and appropriate and clear outputs.

SGI: Automatic clinical subgroup identification in omics datasets

SummaryThe ‘Subgroup Identification’ (SGI) toolbox provides an algorithm to automatically detect clinical subgroups of samples in large-scale omics datasets. It is based on hierarchical clustering trees in combination with a specifically designed association testing and visualization framework that can process an arbitrary number of clinical parameters and outcomes in a systematic fashion. A multi-block extension allows for the simultaneous use of multiple omics datasets on the same samples. In this paper, we describe the functionality of the toolbox and demonstrate an application example on a blood metabolomics dataset with various clinical biochemistry readouts in a type 2 diabetes case-control study.Availability and implementationSGI is an open-source package implemented in R. Package source codes and hands-on tutorials are available at https://github.com/krumsieklab/sgi. The QMdiab metabolomics data is included in the package and can be downloaded from https://doi.org/10.6084/m9.figshare.5904022.

Subgroup identification for differential cardio-respiratory fitness effect on cardiovascular disease risk factors: A model-based recursive partitioning approach

The goal of this study is twofold: i) identification of features associated with three cardiovascular disease (CVD) risk factors, and (ii) identification of subgroups with differential treatment effects. Multivariate analysis is performed to identify the features associated with the CVD risk factors: hypertension, diabetes, and dyslipidemia. For subgroup identification, we applied model-based recursive partitioning approach. This method fits a local model in each subgroup of the population rather than fitting one global model for the whole population. The method starts with a model for the overall effect of treatment and checks whether this effect is equally applicable for all individuals under the study based on parameter instability of M fluctuation test over a set of partitioning variables. The procedure produces a segmented model with a differential effect of cardio-respiratory fitness (CRF) corresponding to each subgroup. The subgroups are linked to predictive factors learned by the recursive partitioning approach. This approach is applied to the data from the Ball State Adult Fitness Program Longitudinal Lifestyle Study (BALL ST), where we considered the level of CRF as a treatment variable. The overall results indicate that CRF is inversely associated with hypertension, diabetes and dyslipidemia. The partitioning factors that are selected are related to these risk factors. The subgroup-specific results indicate that for each subgroup, the chance of hypertension, diabetes and dyslipidemia increases with low CRF.

Subgroup Splits in Diverse Work Teams: Subgroup Perceptions but Not Demographic Faultlines Affect Team Identification and Emotional Exhaustion

We investigate the relationship between (hypothetical) subgroup splits (i.e., faultlines), subjectively perceived subgroups, and team identification and emotional exhaustion. Based on the job demands-resources model and on self-categorization theory, we propose that faultline strength and perceived subgroups negatively affect emotional exhaustion, mediated by team identification. We further propose that subgroup identification moderates the mediation such that subgroup identification compensates low levels of team identification. We tested our hypotheses with a two-wave questionnaire study in a sample of 105 participants from 48 teams from various contexts. We found an effect of perceived subgroups on emotional exhaustion mediated by team identification, but no direct or indirect effect of faultline strength on emotional exhaustion. We also could not find that subgroup identification moderates the effect of team identification on emotional exhaustion. We discuss the need for further research on the link of subgroup splits in work teams and the rise of psychological health issues and derive that measures to prevent burnout should primarily focus on avoiding or reducing subgroup perception whereas affecting the actual demographic composition of work team should be of lower priority.

A high-content platform for physiological profiling and unbiased classification of individual neurons

ABSTRACTHigh-throughput physiological assays often lose single cell resolution, precluding subtype-specific analyses of neuronal activation mechanism and drug effects. Here, we demonstrate APPOINT, Automated Physiological Phenotyping Of Individual Neuronal Types. This physiological assay platform combines calcium imaging, robotic liquid handling, and automated analysis to generate physiological activation profiles of single neurons at a large scale. Using unbiased techniques, we quantify responses to multiple sequential stimuli, enabling subgroup identification by physiology and probing of distinct mechanisms of neuronal activation within subgroups. Using APPOINT, we quantify primary sensory neuron activation by metabotropic receptor agonists and identify potential contributors to pain signaling. Furthermore, we expand the role of neuroimmune interactions by showing that human serum can directly activate sensory neurons, elucidating a new potential pain mechanism. Finally, we apply APPOINT to develop a high-throughput, all-optical approach for quantification of activation threshold and pharmacologically separate the contributions of distinct ion channel subsets to optical activation.