Dataset Information

The Importance of Anti-correlations in Graph Theory Based Classification of Autism Spectrum Disorder.

ABSTRACT: With the release of the multi-site Autism Brain Imaging Data Exchange, many researchers have applied machine learning methods to distinguish between healthy subjects and autistic individuals by using features extracted from resting state functional MRI data. An important part of applying machine learning to this problem is extracting these features. Specifically, whether to include negative correlations between brain region activities as relevant features and how best to define these features. For the second question, the graph theoretical properties of the brain network may provide a reasonable answer. In this study, we investigated the first issue by comparing three different approaches. These included using the positive correlation matrix (comprising only the positive values of the original correlation matrix), the absolute value of the correlation matrix, or the anticorrelation matrix (comprising only the negative correlation values) as the starting point for extracting relevant features using graph theory. We then trained a multi-layer perceptron in a leave-one-site out manner in which the data from a single site was left out as testing data and the model was trained on the data from the other sites. Our results show that on average, using graph features extracted from the anti-correlation matrix led to the highest accuracy and AUC scores. This suggests that anti-correlations should not simply be discarded as they may include useful information that would aid the classification task. We also show that adding the PCA transformation of the original correlation matrix to the feature space leads to an increase in accuracy.

SUBMITTER: Kazeminejad A

PROVIDER: S-EPMC7426475 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Amygdala dysfunction has been implicated in numerous neurodevelopmental disorders, including autism spectrum disorder (ASD). Previous studies in mice and humans, respectively, have linked Pac1r/PAC1R function to social behavior and PTSD-susceptibility. Based on this connection to social and emotional processing and the central role played by the amygdala in ASD, we examined a putative role for PAC1R in social deficits in ASD and determined the pattern of gene expression in the developing mouse and human amygdala. We reveal that Pac1r/PAC1R is expressed in both mouse and human amygdala from mid-neurogenesis through early postnatal stages, critical time points when altered brain trajectories are hypothesized to unfold in ASD. We further find that parents of autistic children carrying a previously identified PTSD-risk genotype (CC) report greater reciprocal social deficits compared to those carrying the non-risk GC genotype. Additionally, by exploring resting-state functional connectivity differences in a subsample of the larger behavioral sample, we find higher functional connectivity between the amygdala and right middle temporal gyrus in individuals with the CC risk genotype. Thus, using multimodal approaches, our data reveal that the amygdala-expressed PAC1R gene may be linked to severity of ASD social phenotype and possible alterations in brain connectivity, therefore potentially acting as a modifier of amygdala-related phenotypes. Autism Res 2019, 12: 200-211 © 2018 International Society for Autism Research, Wiley Periodicals, Inc. LAY SUMMARY: In this multimodal study across mouse and human, we examined expression patterns of Pac1r/PAC1R, a gene implicated in social behavior, and further explored whether a previously identified human PTSD-linked mutation in PAC1R can predict brain connectivity and social deficits in ASD. We find that PAC1R is highly expressed in the both the mouse and human amygdala. Furthermore, our human data suggest that PAC1R genotype is linked to severity of social deficits and functional amygdala connectivity in ASD.

Dataset Information

The Importance of Anti-correlations in Graph Theory Based Classification of Autism Spectrum Disorder.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets