Home

The table below illustrates the performance of ViaComplex for different network sizes. Also, the files available at

“Input_1” and “Input_2” are ready for “Custom Model” analysis, exemplifying the data format.

 

Many other examples are provided within ViaComplex installation package. Please, consider that samples to

exemplify the use of “Processed Model” of genome maintenance mechanisms (GMM).

 

 

Table 1. Numerical examples and corresponding performance of “Custom Model / Functional Analysis” module.

Network graphs a

Input 1: the functional data

Input 2: the network

Performance c

Array samples

Species

Gene IDs

Network attributes

(nodes / links) b

Node IDs

Example_1

 

GSM155194 f

H.sapiens

Gene Symbol

877 / 6497 i

Gene Symbol

1 min 02s

Result 1

Example_2

Different

GSM155194 f

H.sapiens

Gene Symbol

1892 / 26527 i

Gene Symbol

4 min 34s

Result 2

Example_3

network sizesd

GSM155194 f

H.sapiens

Gene Symbol

3837 / 109670 i

Gene Symbol

44 min 53s

Result 3

Example_4

 

GSM155194 f

H.sapiens

Gene Symbol

10000 / 108928 j

Gene Symbol

42 min 15s

Result 4

Example_5

Different organismse

GSM233176 g

S.cerevisiae

ORF ID

4723 / 55930 i

ORF ID

13 min 06s

Result 5

Example_6

GSM174666 h

E.coli K12

ORF ID

3239 / 20284 i

ORF ID

3 min 06

Result 6

Example_7

Hypothetical datad

Hypothetical

Hypothetical

Hypothetical

50 / 500

Hypothetical

0 min 17s

Result 7

 

a The network graphs are shown just for illustrative purposes. They can be visualized in Pajek software using the “Input 2” files

  (http://pajek.imfm.si/doku.php). Please, note that the human GMM network is already available within the software.

b Network data input for “Custom Model / Function Analysis” module.

c Estimated performances in Core 2 Duo 2.00 MHZ 32 bits PC, 3 GB RAM, using <command line> version of ViaComplex V1.0 and default settings.

The <GUI interface> provided essentially the same results for small networks (<3000 nodes); however, the <command line> version gives a better performance for large networks.

d Note that any gene ID can be used, as long as “Input 1” and “Input 2” are provided with same gene/node ID types.

e Likewise, any organism can be analyzed, as long as “Input 1” and “Input 2” are provided with the same identifiers

 (i.e. array and network of the same organism and with the same gene/node ID types).

f Samples public available at GEO database http://www.ncbi.nlm.nih.gov/projects/geo/query/acc.cgi?acc=GSM155194.

g Sample public available at GEO database http://www.ncbi.nlm.nih.gov/projects/geo/query/acc.cgi?acc=GSM233176.

h Sample public available at GEO database http://www.ncbi.nlm.nih.gov/projects/geo/query/acc.cgi?acc=GSM174666.

i These networks were built using the offline version of STRING database (http://string.embl.de/). The online version is able to build only small networks; therefore, we built the networks using the STRING SQL database (free for academic use under request).

j Random links.