Quantitative assessment of biological impact using overlap methods

摘要

Scores for particular network models (those having a source node connected to a set of measurable downstream nodes via causal edges) are computed across multiple networks by accounting for an overlap between these models in a manner that reduces cross-network redundancy and increases the specificity of the network models for the network in which they are found. According to another aspect, a meta-network model is created for networks by accounting for the occurrence of network models that are found in multiple networks in a manner that reduces the redundancy across networks and that increases the specificity of the network model score. Preferably, this process provides additional weighting factors for each node in the network model.

主权项

1. A method to determine a score for a degree of activation of an original network among a set of networks, the original network composed of causal connections among a set of nodes that represent biological entities, processes or other networks, the original network having a reference node, comprising:
(a) constructing a listing representative of a tree graph of tree-depth 1 comprising a root node, and a set of child nodes, wherein the root node of the tree graph represents the original network, and wherein each child node represents a particular node in the original network and is connected to the root node by a signed, directed edge pointing from the root node to that child node; wherein a sign of the signed, directed edge from the root node to that child node is determined from the signs of one or more paths connecting the particular node in the original network to the reference node in the original network; wherein each child node also has a weight associated therewith, the node weight being based on the one or more paths connecting the particular node in the original network to the reference node in the original network; (b) scoring the tree graph based on scores assigned to the child nodes and the signs of the signed, directed edges; and (c) computing a new node weight for each child node by dividing the weight associated with each child node by the number of other networks in the set of networks that also contain that child node, and further normalizing the new node weights to sum to 1 by dividing all new node weights by the sum of all new node weights; wherein at least one of the constructing, scoring and computing steps are implemented in software executing in a hardware element.