Re: [igraph] randomize network with more controls

[Tamas Nepusz]

You remove the original edges and then add the new ones ;) Or, even
better, convert the graph to an adjacency list, do the rewiring there,
and then convert the adjacency list back to a graph.

T.
T.


On Thu, Sep 3, 2015 at 9:41 PM, Qunawei Zhang <address@hidden> wrote:
> Thanks. Is there a function, by which I can rewire two selected edges?
>
> Best
> Quanwei
>
> On 9/3/15 3:38 PM, "Tamas Nepusz" <address@hidden> wrote:
>
>>Hi,
>>
>>There is no built-in function for this in igraph; you'll have to
>>implement the rewiring process yourself.
>>
>>T.
>>T.
>>
>>
>>On Thu, Sep 3, 2015 at 6:25 PM, Qunawei Zhang <address@hidden>
>>wrote:
>>> Hello:
>>>
>>> I know how to use rewire and deSeq to randomize a network and keep the
>>> degree of the nodes.
>>> I have a protein-protein interaction network (192 nodes and 561 edges).
>>> Unlike other protein-protein interaction network, I found the
>>>interactions
>>> among hubs are not suppressed. I want to test whether this topology will
>>> make the subnetwork more vulnerable for random attacks.
>>> So  I need some random network for comparison. To randomize the network
>>>I
>>> want to reduce the interactions among hub genes and on the other hand
>>> increase the number of interactions among hubs and non hubs. Is it
>>>possible
>>> to used some existing method in igraph to get this?
>>>
>>> Many thanks
>>>
>>> Best
>>> Quanwei
>>>
>>>
>>>
>>> _______________________________________________
>>> igraph-help mailing list
>>> address@hidden
>>> https://lists.nongnu.org/mailman/listinfo/igraph-help
>>>
>>
>>_______________________________________________
>>igraph-help mailing list
>>address@hidden
>>https://lists.nongnu.org/mailman/listinfo/igraph-help
>
>
>
> _______________________________________________
> igraph-help mailing list
> address@hidden
> https://lists.nongnu.org/mailman/listinfo/igraph-help