Based on Mantis-ML's predictions it is possible to construct a gene or disease network.
The network is constructed through considering the Mantis-ML scores across all diseases (for a given resource) then finding other genes that are 'close' to these genes based on this vector of scores.
In particular we employ a nonlinear t-SNE embedding of all disease scores and project these down to two dimensions (shown below).
Highlighting a gene reveals its 20 nearest neighbours in the plot. For each of these clusters we provide PheWAS and GO enrichments. PheWAS enrichments were tested against ICD10 codes in UK Biobank at significance level p<10-5.
Highlighting a gene reveals its 20 nearest neighbours in the plot. For each of these clusters we provide PheWAS and GO enrichments. PheWAS enrichments were tested against ICD10 codes in UK Biobank at significance level p<10-5.
Explore the network
Search for a gene:
Search for a disease:
Top cluster:
Rank clusters: