Genetics Computational modeling Flashcards
Imaging genetics
Combines brain imaging and genetics
- to identify and characterize genetic variants (= genotypes = genetic makeup
- associated with inter-individual variation in brain structure and function (=phenotype = observable characteristics)
- to relate this to behavioral traits
Genetic jargon
DNA (deoxyribonucleic acid): spiraling, complex molecule containing base pairs
Gene: length of DNA that holds instructions for making one protein
Chromsome: long strand of DNA, coiled and wrapped up, contains many genes
Genetic variation
Polymophism: variation in a gene or segment of DNA
Single nucleotide polymorphism: variation of single base pair at specific genomic location: some individuals have one nucleotide, others have different nucleotide at that location
Twin studies
Allows determining magnitude of genetic and environmental factors contributing to tair variance
- compare traits resemblance in genetically identical participants with trait resemblance in participants that share only half of their genetic material
Genetic approaches
How to identify genes involved in particular disease/trait/behavior?
Family linkage studies
- standard approach to gene discovery: study families in which disease occurs frequently
- successful in identifying disease-susceptibility genes in rare familial diseases
- mainly successful for diseases caused by mutations in a single gene
- unsuccessful in more common heritable diseases where many genes interact
Candidate-gene approach
Test specific a priori hypotheses regarding the link between measured variation in a gene and variation in brain structure and function
- one or more SNPs for which there is some prior evidence of association with trait or disease are tested against an imaging phenotype
- high statistical power but incapable of discovering new gene combinations
Genome wide association studies
Examination of large number of SNPs across the genome to see if any are associated with specific phenotype
- can pinpoint genes regardless of whether their function was known before but low power owing to the number of independent tests performed
- results can serve as input for candidate-gene approach
Consortia and meta-analytic approaches
To increase statistical power and identify consistent gentic effects
- multicenter consortia combine umerous samples to prduce large-scale datasets
- meta-analytic studies synthesize and combine imaging genetic findings from many cohorts worldwide
Polygenic risk scring
Incorprates effects of all SNPs across the genome by using a single aggregated quantitative metric beased n previously published, publicly available GWA data
Polygenic risk scoring
- theroetical bases
- multiple risk polymorphisms in same disease-related biological pathway will be more liekly to disrupt normal functioning of that pathway
- multiple risk polymorphisms affecting various biological pathways together will predispose or lead to disease
Computational modelign
Use of mathematics, physics and computer science to study behavior of complex systems
- to find mechanistic explanations of how the nervous system processes information
Why build models
- explicit hypotheses and assumptions necessary to test theories of cognition
- provides framework for integrating knowledge from various fields
- allows to observe complex interactions among hypotheses
- provides ultimate control
- leads to empirical predictions
- artificial lesioning possible to test a model’s validity
Computational model based on biological neural netwroks
Different levels of complexity
- from single neurons and synapses up to abstract connectionist-type or population-level descriptions of neural networks
Basic form Computational model
3 different layers
- input layer: receiving information
- hidden layers: processing information
- output layer: transmitting information
Deep neural netwrok Computational model
- Neural network with multiple layers
- between the input and output layers
Biological plausibility of neural network => constraints
- multilayered neural architecture
- each processing node (~ neuron) is connected to many others
- each connection (~ synapse) is characterized by a connection weight –> positive values ~ degrees of excitation; negative values ~ degrees of inhibition
- each simulated neuron receives multiple inputs (~ dendrites)
- if sum of inputs > threshold: output triggered from receiving node (according to a transfer function)
- memory, learning, … depend heavily on changing connection weights (~ synaptic
properties) => ~ experience-dependent plasticity mechanisms
Biophysical models
Simulate behavior of neurons/neural network using biologically inspired mathematical equeations: based on specific assumptions and neurophysiological processes
Multi-modal imaging
Combination of data obtained with different instruments
- combine strengths of different appraoches and overcome weaknesses of each method in isolation
- simultaneous or separate recordings
- integrate information about space, time, brain chemistry, causality
- safety is crucial with additional equipment
- study designs appropriate for both methods
- multiplied degrees of freedom in analyses
Benefits of multi-modal imaging
- improving spatial and temporal resolution: one modality’s superior temporal resolution combined with superior spatial resolution of other modality
- get a more comprehensive physiological view on brain processes
Advantages of simultaneous recording
- mandatory if neuronal events are state-dependent and vary with context
- no between-subjects variances
- no order/practice effects
- identical situation
Disadvantages of simultaneous recording
- specific instrumentation has to be developed
- degraded data quality in terms of signal-to-noise and increased artifacts
- increased subject discomfort and set-up time
Multi-modal imaging in fMRI
+ high spatial resolution, including subcrtical areas
- low temporal resolution, but still allows event-related designs
Multi-modal imaging in EEG
+ high temporal resolution, online monitoring of cogntive processes
- low spatial resolution
- limited to cortical surface
Multi-modal imaging fMRI-EEG
- 2 parallel data sets, analyze separately and compare, correlate etc.
- use fMRI localizer for EEG source reconstruction
- use EEG single-trial amplitude as parametric modulator in GLM
Multi-modal imaging in PET
+ neurotransmitter binding
+ OK spatial resolution
- very low temporal resolution, no event related designs
- rather invasive
- difficult logistics
Multi modal imaging in TMS
+ stimualtion of regions affects ongoing processes
+ rTMS can result in long-lasting changes
- limited to cortical surface
- no direct measure of cortical activity
- somewhat invasive
Pharmacological manipulations
Pharmacological manipulations can be combined with any method under strict ethical guidelines
