Open jlaw9 opened 4 years ago
In the combined
STRING network (cutoff 400):
In the experiments
(i.e., PPIs only) STRING network (cutoff 400):
Thanks! Can you also compute something like the conductance and the average shortest path length?
On Tue, Apr 28, 2020 at 4:16 PM Jeff Law notifications@github.com wrote:
In the combined STRING network (cutoff 400), there is
- 291 Krogan nodes are in one large component
- 37 Krogan nodes in 32 other components (<= 3 Krogan nodes in each)
In the experiments (i.e., PPIs only) STRING network (cutoff 400):
- 135 Krogan nodes are in one component
- 169 Krogan nodes in 159 components (<= 5 Krogan nodes in each)
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/Murali-group/SARS-CoV-2-network-analysis/issues/8#issuecomment-620830829, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABURMUH5SZFHKXW347UFAG3RO42PDANCNFSM4MTCL2HQ .
Here are histograms of the (unweighted) shortest path lengths for the two STRING networks
Hi All - I can confirm Jeff's counts for stringv11 (up to an off-by-one issue). I've also done the same with the tissuenet lung graphs:
stringv11 Graph has 332 nodes and 954 edges
stringv11 Graph has 38 connected components (largest contains 290 nodes)
tissuenet_v2-hpa-protein-lung Graph has 332 nodes and 229 edges
tissuenet_v2-hpa-protein-lung Graph has 202 connected components (largest contains 111 nodes)
tissuenet_v2-gtex-rnaseq-lung Graph has 332 nodes and 361 edges
tissuenet_v2-gtex-rnaseq-lung Graph has 143 connected components (largest contains 183 nodes)
tissuenet_v2-hpa-rnaseq-lung Graph has 332 nodes and 374 edges
tissuenet_v2-hpa-rnaseq-lung Graph has 139 connected components (largest contains 185 nodes)
The degree distribution is similar for the tissuenet induced subgraphs - as expected String is much more connected.
I've also posted the graphs to GS (though I'm having trouble overwriting them - is the GS server working?). They will be posted to the GS Group that Jeff made when I'm done & I'll put them in the slack channel.
We want to know how well connected the human proteins that interact with SARS-CoV-2 proteins from the Krogan dataset are in the PPI network. We can start with the # connected components in the subgraph formed by the Krogan nodes.