Version 0.1.0

RWR Toolkit v0.1.0

This is the initial release of RWRtoolkit. RWRtoolkit enables RandomWalk with Restart (RWR) on both homogenous and heterogeneous multiplex networks. A heterogeneous network is used when integrating multiple network sources; for example in building a multiplex network with a gene-to-gene network and a disease-to-disease networks and combining them by defining a gene-to-disease network which serves as the bi-partite edges. RWRtoolkit provides functions for both creating the muliplex networks and running RWR.

RWRtoolkit has the following methods:

RWR_make_multiplex

Creates an Data object to represent the multiplex network. Generated from an flist. Please see the [README](https://github.com/dkainer/RWRtoolkit/blob/main/README.md#inital-step).

Delta sets the probability to change between layers at the next step. If delta = 0, the particle will always remain in the same layer after a non-restart iteration. On the other hand, if delta = 1, the particle will always change between layers, therefore not following the specific edges of each layer. The default is 0.5. Note delta must be greater than 0 and less than or equal to 1.

Lambda is for heterogeneous networks only. When building a heterogeneous network (i.e. multiple layer groups connected with bipartite links), the walker can jump between layer groups with probability = lambda when it is at a node with a bipartite link. If lambda=1 then walker will oscillate between groups every time it is at a node with a bipartite link. Default is 0.5.

RWR_CV

RWR Cross Validation performs K-fold cross validation on a single gene set, finding the RWR rank of the left-out genes. Can choose between three modes: (1) leave-one-out loo to leave only one gene from the gene set out and find its rank, (2) cross-validation kfold to run k-fold cross-validation for a specified value of k, or (3) singletons singletons to use a single gene as a seed and find the rank of all remaining genes.

• Input: Pre-calculated interaction network (using
  RWR_make_multiplex.R), and a single geneset.
• Output: Table/dataframe with the ranking of each gene in the gene set when
  left out, as well as AUPRC and AUROC curves.

RWR_LOE.R

RWR Lines of Evidence has two possible functions. Given one geneset of seeds, rankings for all other genes in the network will be returned. Given a second geneset of genes to be queried, rankings for just the genes in that geneset will be returned. This can be used to build multiple lines of evidence from the various input networks to relate the two gene sets.

• Input: Pre-calculated interaction network (using
  RWR_make_multiplex), and one or two genesets.
• Output: Table/dataframe with a ranking of non-seed genes (either the rest of the genes in the network if only one input geneset is used, or just the genes in the second geneset if one is provided).

RWR_netscore.R

RWR Net Score performs a network intersect between an input network (network) and a gold truth network (gold), e.g. the GO network. It will score the strength of the intersect with multiple metrics.

• Input:
  • A gold standard network as reference.
  • Another network to compare to the gold standard.
• Output: A table containing multiple metrics including the edge intersect
  between the input network and the gold standard network.

RWR_shortestpaths.R

Find shortest paths between genes in gene sets. Given a single gene set, find the shortest paths between the genes in that gene set. Given two gene sets, find the shortest paths for pairs of genes between gene sets.