Dataset Information

Interpretable Machine Learning Reveals Dissimilarities Between Subtypes of Autism Spectrum Disorder.

ABSTRACT: Autism spectrum disorder (ASD) is a heterogeneous neuropsychiatric disorder with a complex genetic background. Analysis of altered molecular processes in ASD patients requires linear and nonlinear methods that provide interpretable solutions. Interpretable machine learning provides legible models that allow explaining biological mechanisms and support analysis of clinical subgroups. In this work, we investigated several case-control studies of gene expression measurements of ASD individuals. We constructed a rule-based learning model from three independent datasets that we further visualized as a nonlinear gene-gene co-predictive network. To find dissimilarities between ASD subtypes, we scrutinized a topological structure of the network and estimated a centrality distance. Our analysis revealed that autism is the most severe subtype of ASD, while pervasive developmental disorder-not otherwise specified and Asperger syndrome are closely related and milder ASD subtypes. Furthermore, we analyzed the most important ASD-related features that were described in terms of gene co-predictors. Among others, we found a strong co-predictive mechanism between EMC4 and TMEM30A, which may suggest a co-regulation between these genes. The present study demonstrates the potential of applying interpretable machine learning in bioinformatics analyses. Although the proposed methodology was designed for transcriptomics data, it can be applied to other omics disciplines.

SUBMITTER: Garbulowski M

PROVIDER: S-EPMC7946989 | biostudies-literature | 2021

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Interpretable Machine Learning Reveals Dissimilarities Between Subtypes of Autism Spectrum Disorder.

Garbulowski Mateusz M Smolinska Karolina K Diamanti Klev K Pan Gang G Maqbool Khurram K Feuk Lars L Komorowski Jan J

Frontiers in genetics 20210225

Autism spectrum disorder (ASD) is a heterogeneous neuropsychiatric disorder with a complex genetic background. Analysis of altered molecular processes in ASD patients requires linear and nonlinear methods that provide interpretable solutions. Interpretable machine learning provides legible models that allow explaining biological mechanisms and support analysis of clinical subgroups. In this work, we investigated several case-control studies of gene expression measurements of ASD individuals. We ...[more]

PMID: 33719335

Similar Datasets

Project description:BackgroundClinical and etiological varieties remain major obstacles to decompose heterogeneity in autism spectrum disorders (ASD). Recently, neuroimaging raised new hope to identify neurosubtypes of ASD for further understanding the biological mechanisms behind the disorder.MethodsIn this study, brain structural MRI data and clinical measures of 221 male subjects with ASD and 257 healthy controls were selected from 7 independent sites from the Autism Brain Image Data Exchange database (ABIDE). Heterogeneity through discriminative analysis (HYDRA), a recently-proposed semi-supervised clustering method was utilized to divide individuals with ASD into several neurosubtypes by regional volumetric measures of gray matter, white matter, and cerebrospinal fluid. Voxel-wise volume, clinical measures, dynamic resting-state functional magnetic resonance imaging (R-fMRI) measures among different neurosubtypes of ASD were explored. In addition, support vector machine (SVM) model was applied to test whether the neurosubtyping of ASD could improve diagnostic accuracy of ASD.ResultsTwo neurosubtypes of ASD with different voxel-wise volumetric patterns were revealed. The full-scale intelligence quotient (IQ), verbal IQ, Autism Diagnostic Observation Schedule (ADOS) total scores and ADOS severity scores were significantly different between the two neurosubtypes, the total intracranial volume was correlated with performance IQ in Subtype 1 and was correlated with ADOS communication score and ADOS social score in Subtype 2. Compared with Subtype 2, Subtype 1 showed lower dynamic R-fMRI measures, lower dynamic functional architecture stability, higher mean and lower standard deviation (SD) of concordance among dynamic R-fMRI measures in cerebellum. In addition, classification accuracies between ASD neurosubtypes and healthy controls were significantly improved compared with classification accuracy between entire ASD group and healthy controls.LimitationsThe present study excluded female subjects and left-handed subjects, which limited the ability to investigate the associations between these factors and the heterogeneity of ASD.ConclusionsThe two distinct neuroanatomical subtypes of ASD validated by other data modalities not only adds reliability of the result, but also bridges from brain phenomenology to clinical behavior. The current neurosubtypes of ASD could facilitate understanding the neuropathology of this disorder and could be potentially used to improve clinical decision-making process and optimize treatment.

Project description:BackgroundNecroptosis is a novel form of controlled cell death that contributes to the progression of various illnesses. Nonetheless, the function and significance of necroptosis in autism spectrum disorders (ASD) remain unknown and require further investigation.MethodsWe utilized single-nucleus RNA sequencing (snRNA-seq) data to assess the expression patterns of necroptosis in children with autism spectrum disorder (ASD) based on 159 necroptosis-related genes. We identified differentially expressed NRGs and used an unsupervised clustering approach to divide ASD children into distinct molecular subgroups. We also evaluated immunological infiltrations and immune checkpoints using the CIBERSORT algorithm. Characteristic NRGs, identified by the LASSO, RF, and SVM-RFE algorithms, were utilized to construct a risk model. Moreover, functional enrichment, immune infiltration, and CMap analysis were further explored. Additionally, external validation was performed using RT-PCR analysis.ResultsBoth snRNA-seq and bulk transcriptome data demonstrated a greater necroptosis score in ASD children. Among these cell subtypes, excitatory neurons, inhibitory neurons, and endothelials displayed the highest activity of necroptosis. Children with ASD were categorized into two subtypes of necroptosis, and subtype2 exhibited higher immune activity. Four characteristic NRGs (TICAM1, CASP1, CAPN1, and CHMP4A) identified using three machine learning algorithms could predict the onset of ASD. Nomograms, calibration curves, and decision curve analysis (DCA) based on 3-NRG have been shown to have clinical benefit in children with ASD. Furthermore, necroptosis-based riskScore was found to be positively associated with immune activation. Finally, RT-PCR demonstrated differentially expressed of these four NRGs in human peripheral blood samples.ConclusionA comprehensive identification of necroptosis may shed light on the underlying pathogenic process driving ASD onset. The classification of necroptosis subtypes and construction of a necroptosis-related risk model may yield significant insights for the individualized treatment of children with ASD.

Project description:OBJECTIVE:The Centers for Disease Control and Prevention (CDC) coordinates a labor-intensive process to measure the prevalence of autism spectrum disorder (ASD) among children in the United States. Random forests methods have shown promise in speeding up this process, but they lag behind human classification accuracy by about 5%. We explore whether more recently available document classification algorithms can close this gap. MATERIALS AND METHODS:Using data gathered from a single surveillance site, we applied 8 supervised learning algorithms to predict whether children meet the case definition for ASD based solely on the words in their evaluations. We compared the algorithms' performance across 10 random train-test splits of the data, using classification accuracy, F1 score, and number of positive calls to evaluate their potential use for surveillance. RESULTS:Across the 10 train-test cycles, the random forest and support vector machine with Naive Bayes features (NB-SVM) each achieved slightly more than 87% mean accuracy. The NB-SVM produced significantly more false negatives than false positives (P = 0.027), but the random forest did not, making its prevalence estimates very close to the true prevalence in the data. The best-performing neural network performed similarly to the random forest on both measures. DISCUSSION:The random forest performed as well as more recently available models like the NB-SVM and the neural network, and it also produced good prevalence estimates. NB-SVM may not be a good candidate for use in a fully-automated surveillance workflow due to increased false negatives. More sophisticated algorithms, like hierarchical convolutional neural networks, may not be feasible to train due to characteristics of the data. Current algorithms might perform better if the data are abstracted and processed differently and if they take into account information about the children in addition to their evaluations. CONCLUSION:Deep learning models performed similarly to traditional machine learning methods at predicting the clinician-assigned case status for CDC's autism surveillance system. While deep learning methods had limited benefit in this task, they may have applications in other surveillance systems.

Project description:Autism spectrum disorder (ASD) is a highly complex neurodevelopmental disorder characterized by deficits in sociability and repetitive behaviour, however there is a great heterogeneity within other comorbidities that accompany ASD. Recently, gut microbiome has been pointed out as a plausible contributing factor for ASD development as individuals diagnosed with ASD often suffer from intestinal problems and show a differentiated intestinal microbial composition. Nevertheless, gut microbiome studies in ASD rarely agree on the specific bacterial taxa involved in this disorder. Regarding the potential role of gut microbiome in ASD pathophysiology, our aim is to investigate whether there is a set of bacterial taxa relevant for ASD classification by using a sibling-controlled dataset. Additionally, we aim to validate these results across two independent cohorts as several confounding factors, such as lifestyle, influence both ASD and gut microbiome studies. A machine learning approach, recursive ensemble feature selection (REFS), was applied to 16S rRNA gene sequencing data from 117 subjects (60 ASD cases and 57 siblings) identifying 26 bacterial taxa that discriminate ASD cases from controls. The average area under the curve (AUC) of this specific set of bacteria in the sibling-controlled dataset was 81.6%. Moreover, we applied the selected bacterial taxa in a tenfold cross-validation scheme using two independent cohorts (a total of 223 samples-125 ASD cases and 98 controls). We obtained average AUCs of 74.8% and 74%, respectively. Analysis of the gut microbiome using REFS identified a set of bacterial taxa that can be used to predict the ASD status of children in three distinct cohorts with AUC over 80% for the best-performing classifiers. Our results indicate that the gut microbiome has a strong association with ASD and should not be disregarded as a potential target for therapeutic interventions. Furthermore, our work can contribute to use the proposed approach for identifying microbiome signatures across other 16S rRNA gene sequencing datasets.

Dataset Information

Interpretable Machine Learning Reveals Dissimilarities Between Subtypes of Autism Spectrum Disorder.

Publications

Interpretable Machine Learning Reveals Dissimilarities Between Subtypes of Autism Spectrum Disorder.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets