search

meringlab / FlashWeave.jl

Inference of microbial interaction networks from large-scale heterogeneous abundance data
Other
74 stars 8 forks source link

graph() function does not open up #19

Closed dominikmerges closed 4 years ago

dominikmerges commented 4 years ago

Hello,

I'm using FlashWeave with julia-1.5.0. When executing:

data_path = "/home/dmerges/davos_networks/HMP_SRA_gut_small.tsv" netw_results = learn_network(data_path, sensitive=true, heterogeneous=false) graph(netw_results) All I get is: {50, 50} undirected simple Int64 graph with Float64 weights

The same for my own OTu table.

Any ideas are very welcome.

Many thanks and cheers, Dominik

jtackm commented 4 years ago

Hi,

Yes, this is expected. The graph function provides access to the weighted graph object that FlashWeave uses internally to store interactions. This object plays well with the Julia Graph ecosystem (see https://github.com/JuliaGraphs for a collection of packages) and can be used to run a variety of algorithms for downstream analysis.

I assume you wanted to plot the network? FlashWeave does not visualise, you either want to export the network via save_network and load it with a dedicated graph plotting software (for instance Cytoscape, Gephi or one of several python and R packages) or try to use the GraphPlot.jl package directly with the graph object (I haven't used that one in a while, though).

Cheers, Janko

dominikmerges commented 4 years ago

Great, thanks for the fast response!

jtackm commented 4 years ago

In the meantime I fixed the README example, it should be more clear now