Learning debiased graph representations from the OMOP common data model for synthetic data generation.

BACKGROUND: Generating synthetic patient data is crucial for medical research, but common approaches build up on black-box models which do not allow for expert verification or intervention. We propose a highly available method which enables synthetic data generation from real patient records in a privacy preserving and compliant fashion, is interpretable and allows for expert intervention. METHODS: Our approach ties together two established tools in medical informatics, namely OMOP as a data standard for electronic health records and Synthea as a data synthetization method. For this study, data pipelines were built which extract data from OMOP, convert them into time series format, learn temporal rules by 2 statistical algorithms (Markov chain, TARM) and 3 algorithms of causal discovery (DYNOTEARS, J-PCMCI+, LiNGAM) and map the outputs into Synthea graphs. The graphs are evaluated quantitatively by their individual and relative complexity and qualitatively by medical experts. RESULTS: The algorithms were found to learn qualitatively and quantitatively different graph representations. Whereas the Markov chain results in extremely large graphs, TARM, DYNOTEARS, and J-PCMCI+ were found to reduce the data dimension during learning. The MultiGroupDirect LiNGAM algorithm was found to not be applicable to the problem statement at hand. CONCLUSION: Only TARM and DYNOTEARS are practical algorithms for real-world data in this use case. As causal discovery is a method to debias purely statistical relationships, the gradient-based causal discovery algorithm DYNOTEARS was found to be most suitable.

TAOK2 rescues autism-linked developmental deficits in a 16p11.2 microdeletion mouse model.

The precise development of the neocortex is a prerequisite for higher cognitive and associative functions. Despite numerous advances that have been made in understanding neuronal differentiation and cortex development, our knowledge regarding the impact of specific genes associated with neurodevelopmental disorders on these processes is still limited. Here, we show that Taok2, which is encoded in humans within the autism spectrum disorder (ASD) susceptibility locus 16p11.2, is essential for neuronal migration. Overexpression of de novo mutations or rare variants from ASD patients disrupts neuronal migration in an isoform-specific manner. The mutated TAOK2α variants but not the TAOK2ß variants impaired neuronal migration. Moreover, the TAOK2α isoform colocalizes with microtubules. Consequently, neurons lacking Taok2 have unstable microtubules with reduced levels of acetylated tubulin and phosphorylated JNK1. Mice lacking Taok2 develop gross cortical and cortex layering abnormalities. Moreover, acute Taok2 downregulation or Taok2 knockout delayed the migration of upper-layer cortical neurons in mice, and the expression of a constitutively active form of JNK1 rescued these neuronal migration defects. Finally, we report that the brains of the Taok2 KO and 16p11.2 del Het mouse models show striking anatomical similarities and that the heterozygous 16p11.2 microdeletion mouse model displayed reduced levels of phosphorylated JNK1 and neuronal migration deficits, which were ameliorated upon the introduction of TAOK2α in cortical neurons and in the developing cortex of those mice. These results delineate the critical role of TAOK2 in cortical development and its contribution to neurodevelopmental disorders, including ASD.

Altered TAOK2 activity causes autism-related neurodevelopmental and cognitive abnormalities through RhoA signaling.

Atypical brain connectivity is a major contributor to the pathophysiology of neurodevelopmental disorders (NDDs) including autism spectrum disorders (ASDs). TAOK2 is one of several genes in the 16p11.2 microdeletion region, but whether it contributes to NDDs is unknown. We performed behavioral analysis on Taok2 heterozygous (Het) and knockout (KO) mice and found gene dosage-dependent impairments in cognition, anxiety, and social interaction. Taok2 Het and KO mice also have dosage-dependent abnormalities in brain size and neural connectivity in multiple regions, deficits in cortical layering, dendrite and synapse formation, and reduced excitatory neurotransmission. Whole-genome and -exome sequencing of ASD families identified three de novo mutations in TAOK2 and functional analysis in mice and human cells revealed that all the mutations impair protein stability, but they differentially impact kinase activity, dendrite growth, and spine/synapse development. Mechanistically, loss of Taok2 activity causes a reduction in RhoA activation, and pharmacological enhancement of RhoA activity rescues synaptic phenotypes. Together, these data provide evidence that TAOK2 is a neurodevelopmental disorder risk gene and identify RhoA signaling as a mediator of TAOK2-dependent synaptic development.