A network matching algorithm is proposed for matching the de Bruijn graph of contigs against reference genes, to derive 'gene paths' in the graph (sequences of contigs containing gene fragments) that have the highest similarities to known genes, allowing gene fragments contained in multiple contigs to be connected to form more complete (or intact) genes.
Correspondingly, if a R head-to-head and tail-to-tail mirror graphs sequence could be generated from original CR network graph, the multi-path routing problem could be simplified to a single path routing problem in the mirror graphs sequence.
Other examples include: DNA, RNA or protein sequences (linear graphs), sequence fragment overlap graphs (interval graphs) for shotgun sequence assembly, genetic maps and multiple sequence alignments (partial orders).
Two elementary mutation operators, insertion and omission, are exemplarily applied to a hierarchy of graph-based models of increasing expressive power including directed graphs, event sequence graphs, finite-state machines and statecharts.
Currently many drug-target pairs are already known, by which we can take a data-driven approach to search substructure pairs significantly shared in the drug-target (graph-sequence) pairs.
Given significant substructure pairs, for an arbitrary compound-protein (graph-sequence) pair, we can compute a binary vector of 10,000 elements where if a significant substructure pair is included, the value of the corresponding element is 1; otherwise zero.
The scalability of GRASP fingerprints on finding the most similar drug-target pair to an arbitrary given compound-protein (graph-sequence) pair was examined by generating 975,243,103 compound-protein pairs (which we call MASS) from 140,937 bioactive compounds and 6,919 druggable proteins (Methods section and Methods S1).
Our algorithm has two key features: 1) Listing up all frequent substructure pairs (Fig. 1b): This is a mathematical issue of enumerating all frequent pairs of subgraphs and subsequences which appeared in more than a pre-specified percentage (which is called support) in given graph-sequence pairs.
The internal schema of the BiologicalNetworks database is shown in Figure 2. Four orthogonal types of biological data--graphs, sequences, histograms, and tree structures--are integrated to enable multi-scale data analysis for the host-pathogen studies.
The properties of the sequenced genome and quality of the input data is reflected by the structure of the graph; repeats, sequence variation (in a diploid or polyploid genome) and sequencing errors cause branches in the graph.
