interface of bio3d and wisp

[Former user]

I am bit unclear about the interface of protein structure network analysis and wisp as suggested in the vignette.

write.table(net$cij, quote=FALSE, row.names=FALSE, col.names=FALSE, file="adj.txt") File "adj.txt" corresponds to the "functionalized matrix"?

Is this the only information that needs to be exported from bio3d to wisp or covariance matrix and contact map files generated in bio3d should also be supplied? I would be great if all this analysis could be done in bio3d itself!

Also HTML link to the tutorial on the Documentation page is broken:

http://thegrantlab.org/bio3d/tutorials/correlation-network-analysis

[Xinqiu Yao]

Hi,

The command you used is correct; The generated "adj.txt" file should be the input of WISP specified by "-user_specified_functionalized_matrix_filename".

Note that ways to build a network are diverse. The method we employed for our own work is a bit different from that used in the paper of WISP (Van Wart, et al., JCTC2014), e.g. we don't use contact map as extra filter (and so "-contact_map_distance_limit" was set to a large value like 999999.999). Instead, distance information was incorporated before network construction (See the bio3d function filter.dccm() for more details). However, you can always choose your favorite method; either bio3d or wisp doesn't have any restriction on it.

I also delightedly to inform you that we DO have our own function for path analysis, which we added just recently: See help(cnapath) for details. The method implemented in bio3d gives the same results as WISP, but is much faster in some cases. To use these functions, download and install development version bio3d (See the guideline).

We are very sorry about the broken of the link: we have replaced that tutorial with a more comprehensive and user-friendly one. Please update your bookmark with: http://thegrantlab.org/bio3d/tutorials/protein-structure-networks, or go to http://thegrantlab.org/bio3d/tutorials to check out latest changes.

[Xinqiu Yao]

User's questions have been answered.