The workflow analyses data from DrugBank (http://www.drugbank.ca/) a public available db that combines detailed drug (i.e. chemical, pharmacological and pharmaceutical) data with comprehensive drug target (i.e. sequence, structure, and pathway) information.
Each drug is described by a drug card with more than 150 data fields of which only some are used in this workflow for demonstration.
All necessary files are available at www.drugbank.ca/downloads
The workflow demonstrates how the partition information can be used during network analysis.
It analyses schools Wikipedia (http://schools-wikipedia.org) using the network analysis plugin.
Schools Wikipedia is a subset of the English Wikipedia (http://en.wikipedia.org) and contains more than 5000 articles.
Articles are represented as nodes in the network with the title of the article as node id.
This workflow demonstrates the filtering of a network.
It demonstrates the filtering of nodes based on a given id list or based on a feature value.
It also demonstrates the filtering of leafs and unconnected nodes as well as the filtering of empty edges and loops.
The workflow demonstrates some basic network analysis tasks such as network filtering, shortest path and connected component detection.
It analyses schools Wikipedia (http://schools-wikipedia.org) using the network analysis plugin.
Schools Wikipedia is a subset of the English Wikipedia (http://en.wikipedia.org) and contains more than 5000 articles.
Articles are represented as nodes in the network with the title of the article as node id.
This workflow predicts unknown feautures using the build in KNIME data mining nodes.
It builds a decision tree model to predict the phase of life attribute for the nodes that miss these information and adds the predicted value to the correspnding nodes.
This workflow demonstrates how to process subgraphs analysis in a loop.
It extracts the direct neighborhood of each node in the network and performes some basic network analysis tasks on each of the subgraphs.
This workflow demonstrates the creation of a network based on an edge table.
It also demonstrates the insertion of node features from an feature table and the insertion of graph features.
It also provides a first view of the network using the network visualizer node.
The workflow clusters the node adjacency matrix and inserts the clustering as new node partitions into the network.
Once all cluster members are assigned to the corresponding cluster partition a partition graph is generated which shows the connections between partitions and the size of each partition.
This workflow finds the most important authors for a new indication via network mining approaches and based on a set of PubMed articles that includes relevant terms of this new indication.
A network with authors as nodes and publications as edges is generated. The number of common publications is used as an edge weight.