Acquisition and analysis of in vivo markers of ‘Excitation and Inhibition’ in humans

Glutmatergic and GABAergic genes are genes that are known to encode proteins involved in glutamate and GABA pathways. By using genome data, competitive gene-set analysis can be performed using MAGMA, an openly available gene and gene-set analysis tool (https://ctg.cncr.nl/software/magma), for investigating associations between genetic variation in glutamate and GABA gene-sets with both behavioral and neuroimaging measures.

Additionally, previous work has been done performing ‘virtual histology’ analyses to investigate correlations between gene expression of genes associated with different cell types in the brain, to cortical thickness, using gene-expression data from the open source allen human brain atlas (https://portal.brain-map.org/explore/overview?gclid=Cj0KCQiA_P6dBhD1ARIsAAGI7HAAbX1HklUv8VmOHeWLxgFfEmVh3KOCRozt_kwfXF-GnMk1S5SKJioaAi6tEALw_wcB). This method can also be implemented using the glutamate and GABA gene-sets, investigating correlations between expression of these genes and cortical thickness.

In this session I will discuss performing and interpreting these analyses. Firstly, competitive gene-set analysis using MAGMA with behavioral and neuroimaging (MRI, EEG) measures. Secondly, the gene-expression associations with cortical thickness and how it can be used to investigate correlations of gene-expression and cortical thickness differences between groups. Lastly I will discuss the benefits and limitations of these analysis methods and show examples of ways these have been implemented to additionally compare groups of study participants.

Learning aims:
1. Understand how competitive gene-set analysis can be performed
2. Understand how gene-expression analysis can be performed
3. Understand how these analysis methods can be used to investigate multimodal links of genetic, behavioral and neuroimaging measures 

Normal brain functioning requires balanced neuronal excitation and inhibition. An imbalance of excitation and inhibition ratio (E/I ratio) has been linked to an array of psychiatric disorders, including autism spectrum disorder and schizophrenia. Thus, characterizing neuronal excitation and inhibition is an important research question to advance our understanding of disease mechanisms and a potential gateway to future therapeutic interventions. However, precise measurements of excitation and inhibition require invasive approaches, which are infeasible for human participants.

In this talk, I will introduce a collection of promising computational modeling approaches for estimating E/I ratio from resting state fMRI. These approaches include large-scale biophysical models, time series analysis and other data-driven methods. I will compare the advantages and limitations of each method. Finally, I will discuss questions of how E/I ratio shifts across development as well as how E/I ratio relates to psychopathology and cognition during brain maturation.

Learning aims:
1. Learn different approaches to estimate E/I ratio using fMRI data
2. Understand the advantages and limitations of each approach
3. Understand the behavioral and psychopathological relevance of E/I ratio 

Pharmacological functional MRI (ph-fMRI) involves administering a pharmacological agent (drug) and acquiring fMRI data while the acute effects of the drug are present. This allows for an empirical assessment of the effect of a drug on the fMRI signal. If a drug is used that can pharmacologically manipulate the excitation/inhibition (E/I) ratio, ph-fMRI can be leveraged to study the association between E/I ratio and patterns of fMRI signal. I will present a ph-fMRI approach that uses a GABAergic benzodiazepine drug challenge to generate a biomarker for a reduced E/I ratio. First, using data from a double-blind, placebo-controlled ph-fMRI dataset, I will show how a putative E/I biomarker can be generated by training a machine learning classifier to distinguish benzodiazepine and placebo sessions based on functional connectivity patterns. Next, I will show how the biological and pharmacological validity of the E/I biomarker can be assessed by comparing the learned classification features to gene expression patterns for benzodiazepine-sensitive GABAA receptors. Finally, I will show an example of how this E/I biomarker can be applied to a new dataset to gain insight into the neurodevelopment of the E/I ratio during youth.

Learning aims:
1. Understand how pharmacological challenge can be used to manipulate the E/I ratio
2. Learn approaches for generating and validating E/I-sensitive machine learning biomarkers from pharmacological fMRI data
3. Understand the potential to apply biomarkers generated from pharmacological fMRI data to new datasets to investigate E/I neurobiology 

Magnetic Resonance Spectroscopy (MRS) allows non-invasive quantification of both GABA and glutamate within the brain. MRS has been used by many groups around the world to understand the relationship between E/I ratio and various aspects of behaviour, in both health and disease. Here I will explain the basis of the MRS signal, highlighting the major advantages and pitfalls of the technique. I will discuss the broad principals of the available analysis approaches, and what the MRS signals may reflect.

I will go on to review the combination of MRS with behaviour to study the neurochemical dynamics of plasticity across behaviours ranging from motor control, visual acuity and memory formation. I will review the evidence for changes in E/I balance (acquired using MRS) in a range of neurological and psychiatric disorders. Finally, I will discuss the role of multimodal studies, including pharmaco-MRS and combination with non-invasive brain stimulation (NIBS) to allow causal inference.

Learning aims:
1. Understand the origin of the MRS-GABA and MRS-glutamate signals
2. Understand how changes in GABA can inform us about brain function and plasticity
3. Understand how MRS of GABA can play an important role in understanding neurological and psychiatric disorders