Network Topology - Transitive Alignments

Transitive Network Topology¶

Transitive Alignment Workflow enhances the construction of Molecular Networks (MNs) in mass spectrometry data analysis by addressing limitations in traditional MS/MS peak alignment methods. Conventional approaches, which rely on pairwise comparisons using modified cosine methods, often fail to connect molecules with multiple structural modifications, resulting in incomplete networks. The Transitive Alignments approach incorporated in this workflow utilizes network topology to connect molecules that are indirectly related, thereby improving the accuracy and completeness of MNs.

Checkout the workflow here.

You can also use this after any molecular networking job, either classical or FBMN.

Citation - In Press

Network Topology Parameters¶

Transitive Param - Min Clique Cosine¶

The score threshold to control the edge score within a CAST clique.

Transitive Param - Min Transitive Alignment Score¶

The score threshold for adding a transitive alignment edge back to the network.

Filter option for each compound after constructing the MN (This is only for better visualization layout). Available options: - Pure MST: Only apply MST on the MN. - Greedy MST: After applying MST, greedily add back high score edges for higher average edge score. - Hybrid MST: Using the original edges first, if unable to create MST, then use the transitive alignment edges to finish the MST.

Induced Network Parameters¶

Induced Network Param - Option¶

Intersection the Transitive and MAX Hops: Using this method, if and only if the nodes in the original network are within the max hops of the source node and also the transitive alignment score is above the setting threshold (Induced Network Param - Min Transitive Alignment Score) will be in the induced network results.
Union the Transitive and Max Hops: This method will first do all the transitive alignment from the source node to all the other reachable nodes, and then add back the transitive alignment edge above the threshold (Induced Network Param - Min Transitive Alignment Score) to the original network. Afterward, select the node within the max hops from the source node or the edge score above the threshold (Induced Network Param - Min Transitive Alignment Score) to the results.

Induced Network Param - Source Node¶

The source node to start the induced network.

Induced Network Param - Max Hops¶

Maximum hops from the source node to the target node.

Induced Network Param - Min Transitive Alignment Score¶

The transitive alignment score threshold.

Filter option for the induced network after constructing (This is only for better visualization layout). Available options: - Pure MST: Only apply MST on the MN. - Greedy MST: After applying MST, greedily add back high score edges for higher average edge score. - Hybrid MST: Using the original edges first, if unable to create MST, then use the transitive alignment edges to finish the MST.