Transcript Introduction to The Stanley Neuropathology Consortium

The Stanley Neuropathology Consortium Integrative Database: A
novel web-based tool for exploring neuropathological traits, gene
expression and associated biological processes in psychiatric disorders
What is the Stanley Neuropathology Consortium Integrative Database (SNCID)
To facilitate psychiatric research through data sharing, the SMRI has developed a
new web-based database.
The SNCID is freely available to all users (http://www.stanleyresearch.org/sncid/).
However, for commercial use, agreement and permission should be obtained from
the SMRI.
The SNCID currently integrates 1747 neuropathology datasets and three
expression microarray datasets with statistical modules.
All data sets used in the SNCID have been generated with the Stanley
Neuropathology Consortium (SNC) tissues.
Descriptive statistical and variance analysis modules enable users to identify
neuropathological traits in multiple brain areas for psychiatric disorders such as
bipolar disorder, depression and schizophrenia.
Integration of genome-wide expression microarray data with correlation modules
allow users to explore genes correlated with the trait of interest. An interface that
links the SNCID to the DAVID database (http://david.abcc.ncifcrf.gov/) allows for the
functional annotation of those probe sets that correlated with a particular trait.
The Stanley Neuropathology Consortium
The Stanley Neuropathology Consortium is a collection of 60
brains, consisting of 15 each diagnosed with schizophrenia, bipolar
disorder, or major depression, and unaffected controls. The four
groups are matched by age, sex, race, postmortem interval, pH,
side of brain, and mRNA quality.
 All data used in the SNCID were generated with tissues from the
SNC.
Overall structure of the SNCID
External
Database and web tool
Repository database
Neuropathology database
Zipped raw data files
12 brain regions
Statistical analysis module
Descriptive analysis
Variance analysis
Correlation analysis
NCBI
DAVID
Microarray database
Frontal cortex
Cerebellum
Application 1: Identification of Neuropathological Trait
Step1-1: Search for trait (marker) of interest
Select brain region of interest (12 regions available) and marker type (RNA, Protein,
Cell or other)
Application 1: Identification of Neuropathalogical Trait
Step1-2: Search trait of interest
Refine search by selecting specific marker and/or researcher
Application 1: Identification of Neuropathology Trait
Step 2: Basic information of marker, study, and related reference
Hyperlink to NCBI’s Entrez gene db
Study information
e.g. methodology
Published paper related to the study
Application 1: Identification of Neuropathology Trait
Step 3: Descriptive statistics
To see the basic statistical information regarding BDNF mRNA levels in
layer VIa of temporal cortex choose ‘Analysis’.
Application 1: Identification of Neuropathology Trait
Step 3: Descriptive statistics with histogram of the marker
‘Analysis’ gives the basic statistics regarding the dataset, number of cases
included (e.g. count) and indicates if the data is normally distributed.
Application 1: Identification of Abnormal Neuropathological Trait
Step 4: Statistical analysis: Variance analysis
To determine if there is a difference in BDNF mRNA levels between diagnostic groups and
controls ‘select a method’ e. g. ANOVA or Non-parametric test.
Application 1: Identification of Abnormal Neuropathology Trait
Step 4: Results of variance analysis with box plot
Non-parametric test gives basic statistical description of group differences.
Application 1: Identification of Abnormal Neuropathology Trait
Step 4: Statistical analysis: Identification of confounding factors
To determine which confounding factor (e.g. select brain pH) correlates (e.g. select Parametric or
Non-parametric) with BDNF mRNA levels in temporal cortex
Application 1: Identification of Abnormal Neuropathology Trait
Step 4: Statistical analysis: Identification of confounding factors
Non-parametric test indicates brain pH is significantly correlated with BDNF
mRNA levels in temporal cortex
Application 1: Identification of Abnormal Neuropathology Trait
Step 4: Statistical analysis: correlation analysis
To determine which markers in another brain area (e.g. select frontal cortex) correlate
(e.g. select Spearman’s) with BDNF mRNA levels in temporal cortex.
Application 1: Identification of Abnormal Neuropathology Traits
Step 4: Statistical analysis: correlation analysis
Spearman’s test reveals 47 markers in frontal cortex are significantly correlated
(p<0.01) with BDNF mRNA levels in temporal cortex
Application 2: Exploration of genes and pathways associated with dopamine
levels in frontal cortex
Variance analysis – reveals dopamine levels are significantly increased in the frontal cortex
of the schizophrenia group as compared to the control group
Application 2: Exploration of genes and pathways associated with dopamine
levels in frontal cortex
Genome-wide correlation analysis
To determine which genes are correlated with dopamine levels in frontal cortex select microarray
dataset (e.g. frontal, MAS5) and method (e.g. Spearman’s)
Application 2: Exploration of genes and pathways associated with dopamine
levels in frontal cortex
Functional annotation interface
Spearman’s test revealed 43 genes significantly correlated (p<0.001) with dopamine levels
in frontal cortex.
To identify the functional pathways associated with these 43 genes choose ‘functional annotation’
Application 2: Exploration of genes and pathways associated with dopamine
levels in frontal cortex
Functional annotation interface
43 genes are uploaded onto the DAVID by functional annotation interface and then select functional
annotation tool in the DAVID
Application 2: Exploration of genes and pathways associated with dopamine
levels in frontal cortex
Functional annotation using DAVID database
Application 3: Exploration of genes and pathways associated with glutamate
levels in frontal cortex
Variance analysis – revealed glutamate levels are significantly elevated in the frontal cortex
of groups with bipolar disorder and depression as compared to the normal control group.
Application 3: Exploration of genes and pathways associated with glutamate
levels in frontal cortex
Confounding factor analysis – reveals a significant difference in glutamate levels between
cases that commit suicide and those that do not.
suicide
Application 3: Exploration of genes and pathways associated with glutamate
levels in frontal cortex
Genome-wide correlation analysis
suicide
151 genes significantly correlated (p<0.01) with glutamate levels in the frontal cortex
Application 3: Exploration of genes and pathways associated with glutamate
levels in frontal cortex
Functional annotation using DAVID database
Repository database
The SNCID also integrates repository database. We strongly recommend users to
download the raw data for further statistical analysis.
Any questions, suggestions, or comments on the SNCID?
We hope our new database will contribute to the identification of novel neuropathological
markers for psychiatric disorders. We hope that the potential mechanisms that underlie
the abnormalities of these markers in psychiatric disorders will be revealed Eventually
we hope the SNCID will help researchers find novel drug targets for the major psychiatric
disorders.
Enjoy data-mining!
Please contact Sanghyeon Kim ([email protected]) or Maree Webster
([email protected]), if you have any questions, suggestions or comments
on the SNCID.