Genetics: Transcriptomics, Phenotypes and Disorders

The Allen Human Brain Atlas (AHBA) consists of gene expression profiled within 3702 tissue samples from six donor brains. The provision of sample coordinates within MNI-space has made comparing neuroimaging findings to normative gene expression feasible and popular; the AHBA has been featured in at least 202 studies over the last five years (via PubMed) and ~44 OHBM 2024 abstracts. In fact, there exist multiple sets of coordinates describing these sample locations: 1) "original" coordinates from the AHBA (Hawrylycz, 2012), 2) updated coordinates from the "alleninf" package (Gorgolewski, 2014) and used in the abagen software (Markello, 2021), and 3) "CIC" coordinates derived at the Cerebral Imaging Centre via multispectral image registrations to a newer MNI template (Devenyi, 2018). Surprisingly, these coordinates place many tissue samples in dramatically different anatomical locations (Figure 1a,b). Here, we test the accuracy of these three coordinate sets through multiple types of tests of location accuracy, and then show that inaccuracies in coordinates can result in improper inferences in neuroimaging studies. 

Graph metrics (Rubinov & Sporns, 2010) derived from the resting-state network and asymmetry thereof have both produced reliable correlates of various neuropsychiatric disorders. However, these phenotypic correlates do not mirror those identified from structural MRI, raising the question if different biological processes underlie brain structure and function, and if they have different clinical correlates. Initial results have shown genetic correlations between the resting-state network and neuropsychiatric disorders (Bell et al., 2022; Roelfs et al., 2023; Zhao et al., 2022), but these results only covered sporadic phenotypes. Hence, we conducted a systematic genome-wide association study (GWAS) across six graph metrics at global, hemispheric and regional levels and their asymmetry (fig. 1), and identified their shared genetics with other imaging-derived phenotypes (IDPs) and 11 neuropsychiatric disorders (fig. 2). 

The human cortical architecture exhibits complex functional divisions and organizational patterns, which are often altered in various psychiatric disorders, with these changes being closely regulated at the molecular level. Understanding the molecular mechanisms underlying cortical organizational patterns relies on transcriptomic data at the whole-cortex scale. Historically, the Allen Human Brain Atlas (AHBA) has been used to correlate different macroscopic networks, deviant patterns, and gene expression, providing insights into the molecular mechanisms. However, due to the limitations of microarray sequencing, while the AHBA data provides transcriptomic information at whole-cortex spatial resolution, it is constrained by sample size and lacks information on specific cell types. Recently, the release of human single-nucleus datasets has partly addressed these shortcomings, though due to sampling constraints, these datasets do not provide full cortical coverage across brain regions. Therefore, combining the strengths of both datasets can deepen our understanding of the complex molecular mechanisms underlying the brain cortex. A key question is whether data from different modalities can represent similar cortical transcriptional patterns. In this study, we analyzed the consistencies and differences between two modalities of data in representing cortical transcriptional patterns. 

Psychiatric disorders often emerge during adolescence and early adulthood, a critical neurodevelopmental period characterized by heightened vulnerability to environmental influences. Early identification of individual risk profiles and timely interventions are therefore essential to improve long-term psychosocial and occupational functioning. However, the heterogeneity of risk and protective factors and lack of reliable biomarkers complicate this task. Neurodevelopmental pathways, driven by genetic predisposition and environmental stressors, contribute to clinical symptoms, cognitive deficits, and psychosocial dysfunction.
This study introduces a novel unsupervised machine learning method, multiblock sparse partial least squares (MB-SPLS), to integrate genetic, neuroanatomical, and clinical data. The multimodal approach provides a more comprehensive understanding of the multi-layered complexity of early-stage psychiatric disorders. Here, we aim to identify clinically relevant risk signatures to predict longitudinal functioning trajectories. 

Suicide in bipolar disorder (BD) with high heritability (Turecki, G., 2019; Erlangsen, A., 2020) is characterized as a dysconnectivity syndrome (Wang H., 2022; Wang, H., 2020) 3-4. While traditionally focused on single-level biological data, suicide-related imaging-genetic studies are now shifting towards a multidimensional approach to spatial correspondence (Li, J., 2024; Qin, K., 2024). It poses challenges for the exploration of multifaceted reliable genetic landscape responsible for reproducible neuroanatomical alterations by suicidal effects. Beyond the exploration of macro-micro-coupling, more tangible and compelling proof-of-principle with paired measures of multi-biological characteristics is also urgently needed to develop personalized risk assessments in clinic (Wang, J., 2024). 

The strongest common genetic association with schizophrenia is linked to increased expression of the complement component 4A (C4A) gene. In the brain, C4A expression is thought to influence synaptic pruning and to act in a sex-specific manner, contributing to a higher risk of schizophrenia in males compared to females. We previously demonstrated in a cohort of ~7,500 youth from the ABCD Study that increased genetically predicted C4 gene expression is associated with reduced surface area of the entorhinal cortex at 9–10 years of age, which predicts greater number and severity of psychosis-like experiences 1–2 years later (Hernandez, 2023). However, the impact of C4A expression on the rate of change (ROC) in brain structure during childhood and adolescence remains unclear. To address this, we investigated the relationship between genetically predicted C4A expression and longitudinal changes in regional brain surface area, cortical thickness, and volume in 2,977 ABCD participants with structural MRI data collected at three timepoints (ages 9–10, 11–12, and 13–14 years). Additionally, we conducted a phenome-wide association study (PheWAS) to evaluate associations between brain regions influenced by C4A expression and 138 psychiatric and behavioral traits assessed at 13–14 years.