Trait Co-occurrence
About This Analysis
Explores how phenotypes co-occur across the gene dataset using Pearson correlation. Traits that frequently appear together may share underlying biological mechanisms. Use the cross-tabulation view to examine specific trait pairs, or the heatmap for an overview of all relationships.
Trait Pairs
-
Possible combinations
Strong Correlations
-
|r| > 0.3
Max Correlation
-
Strongest association
Selected Trait
None
Click to highlight
Trait Correlation Matrix
Click on any cell or axis label to highlight a trait. Blue = positive correlation, Red = negative.
What you're seeing: A heatmap showing Pearson correlations between all trait pairs.
Blue cells indicate traits that tend to co-occur; red cells indicate inverse relationships.
What it means: Strongly correlated traits may share underlying biological mechanisms or represent related aspects of ASD presentation. Click any cell to explore that trait's associations.
What it means: Strongly correlated traits may share underlying biological mechanisms or represent related aspects of ASD presentation. Click any cell to explore that trait's associations.
Correlation Matrix (Circle View)
Circle size and color show correlation strength. Upper triangle: values, Lower triangle: circles. Click traits to explore.
What you're seeing: An alternative correlation view with circles in the lower triangle (size = strength) and numeric values in the upper triangle.
Traits are sorted by correlation strength.
What it means: This view makes it easier to spot strong correlations at a glance. Larger, more saturated circles indicate stronger associations.
What it means: This view makes it easier to spot strong correlations at a glance. Larger, more saturated circles indicate stronger associations.
Trait Co-occurrence Network
Nodes are traits, edges show correlations > 0.25. Node size = number of connections. Drag to rearrange, scroll to zoom.
What you're seeing: A force-directed network where each node is a trait and edges connect correlated traits.
Node size reflects how many other traits each trait correlates with; colors indicate phenotype category.
What it means: Highly connected traits (large nodes) are central to the phenotype landscape and may represent core ASD features. Clusters of connected traits suggest related phenotype domains.
What it means: Highly connected traits (large nodes) are central to the phenotype landscape and may represent core ASD features. Clusters of connected traits suggest related phenotype domains.
Click traits in the table below to add them
Filtered Correlation Heatmap
Showing all traitsTop Trait Associations
Showing top associations
What you're seeing: Trait pairs ranked by how often they appear together in the same genes.
Correlation values range from -1 (never together) to +1 (always together).
What it means: High positive correlations suggest traits that tend to co-occur, which may indicate shared biological mechanisms or overlapping clinical presentations.
What it means: High positive correlations suggest traits that tend to co-occur, which may indicate shared biological mechanisms or overlapping clinical presentations.
| Trait 1 | Trait 2 | Genes with Both | Correlation | Strength |
|---|
Association Rules Mining
Discovering "IF Trait A AND Trait B → Trait C" rules using the Apriori algorithm
What you're seeing: Association rules reveal patterns like "genes with traits A AND B often have trait C."
Support = frequency of rule, Confidence = P(consequent|antecedent), Lift = how much more likely than random chance.
What it means: Rules with lift > 1 indicate positive associations. High-lift rules reveal non-obvious trait combinations that frequently co-occur.
What it means: Rules with lift > 1 indicate positive associations. High-lift rules reveal non-obvious trait combinations that frequently co-occur.
Total Rules
-
Discovered patterns
High-Lift Rules
-
Lift > 1.5
Max Lift
-
Strongest association
Avg Confidence
-
Rule reliability
Top Association Rules by Lift
Rules Network (Antecedent → Consequent)
Directed edges show rule relationships. Node size = frequency in rules.Rule Metrics Comparison
Each line is a rule. Hover to see details.Detailed Rules Table
-| Antecedent(s) | → | Consequent(s) | Support | Confidence | Lift |
|---|