Allen Brain Atlas & KEGG: Part II

The code below finds the KEGG pathways that each gene is linked to (Figures 1 & 2).

Figure 1: Pathways & Genes.  The set of autism associated genes (Allen Brain Institute) associated with KEGG's the signal transduction pathways.

Figure 2: Pathways & Genes.  The set of autism associated genes (Allen Brain Institute) associated with KEGG's signalling molecules and interaction pathways.

Note: Please request the "pathwaysStructure" matrix file from the author.  "fileName" is the name of the batch data file after running the script of Part I.

load(fileName)
load('pathwaysStructure')

% Structures
uniqueStructureID = unique(partsData(:,2));
[structureAbbreviations, uniqueStructureNames] = getStructureName(uniqueStructureID);
uniqueTopStructureID = unique(partsData(:,1));
[topStructureAbbreviations, uniqueTopStructureNames] = getStructureName(uniqueTopStructureID);

% Pathway Group
for j = 1:1:size(Description,1)   
    disp([num2str(j), ': ', Description{j,1}, '-->'])
    disp(Description{j,2})   
end
pathwayGroupID = input('Which Pathway Group (above) are you interested in? ');

% Genes

genesID = [(1:size(geneData(:,1),1))' cell2mat(geneData(:,1))];
genesNames = geneData(:,4);

for n = 1:1:size( Description{pathwayGroupID, 2}, 1 )   
    applicableGenes = [];
    applicablePathways = [];   
    pathwaysTicks = [num2cell((1:size( Description{pathwayGroupID, n + 2}, 1 ))') ...
        Description{pathwayGroupID, n + 2}(:,1)];
    pathwaysNames = Description{pathwayGroupID, n + 2}(:,2);

 
    for k = 1:1:size(pathwaysNames)       
        [startAt, endAt] = regexp(pathwaysNames{k}, ' signaling pathway');       
        if ~isempty([startAt endAt])
           pathwaysNames{k}(startAt:endAt) = '';
        end

     
        [startAt, endAt] = regexp(pathwaysNames{k}, ' signaling system');       
        if ~isempty([startAt endAt])
           pathwaysNames{k}(startAt:endAt) = '';
        end

    
        Index = strfind(pathwaysNames{k}, ' ');
        if length(Index) > 2
           pathwaysNames{k} = pathwaysNames{k}(1:Index(2));
           pathwaysNames{k}(Index(1) + 1) = upper(pathwaysNames{k}(Index(1) + 1));
        end       
    end

 
    figure
    hold on
    for k = 1:1:size( Description{pathwayGroupID, n + 2}, 1 )   
        Index = strcmp( Description{pathwayGroupID, n + 2}(k,1), pathwaysData(:,2));       
        if sum(Index) == 0
            continue;
        end       
        Set = pathwaysData(Index,:);
        iGenes = dsearchn(genesID(:,2), cell2mat(Set(:,1)));
        iPathways = cell2mat(pathwaysTicks( strcmp( Description{pathwayGroupID, n + 2}(k,1), ...
            pathwaysTicks(:,2) ), 1 ));               
        plot(genesID(iGenes,1), iPathways*ones(size(iGenes)), 'o', 'MarkerFaceColor', rand(1,3), 'MarkerEdgeColor', 'none')   
        applicableGenes = [applicableGenes; genesID(iGenes,1)];       
        applicablePathways = [applicablePathways; iPathways];       
    end   
    if isempty(applicableGenes)
        continue;
    end   
    applicableGenes = unique(applicableGenes);
    ylim([0 size(pathwaysTicks,1) + 1])
    xlim([min(applicableGenes) - 1 max(applicableGenes) + 1])  
    set(gca, 'XTick', applicableGenes, 'XTickLabel', genesNames(applicableGenes), ...
        'YTick', cell2mat(pathwaysTicks(:,1)), 'YTickLabel', pathwaysNames)
    xticklabel_rotate   
    box on   
    xlabel('Genes')
    ylabel([Description{pathwayGroupID, 2}(n), ' Pathways'])
    title({ [Description{pathwayGroupID, 1}, ...
        ' Pathways Group']; [Description{pathwayGroupID, 2}{n,1}, ' Pathways'] })
end