https://github.com/arnikz/netclust

Efficient cluster detection tool for large graphs (networks)
https://github.com/arnikz/netclust

bioinformatics cluster-detection graph-algorithms large-dataset

Last synced: 3 months ago
JSON representation

Efficient cluster detection tool for large graphs (networks)

• Host: GitHub
• URL: https://github.com/arnikz/netclust
• Owner: arnikz
• License: gpl-3.0
• Created: 2015-09-16T10:47:10.000Z (almost 9 years ago)
• Default Branch: master
• Last Pushed: 2018-07-27T02:24:36.000Z (almost 6 years ago)
• Last Synced: 2024-01-22T12:00:55.022Z (5 months ago)
• Topics: bioinformatics, cluster-detection, graph-algorithms, large-dataset
• Language: C
• Homepage:
• Size: 56.6 KB
• Stars: 1
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Lists

• awesome-community-detection - [C Reference

README

        
netclust/netindex

=================

are fast and memory-efficient C programs developed to detect clusters of nodes in very large graphs (networks) with billions of edges using a scalable nearest-neighbor approach with (nearly) linear time and space complexity *~O(E)*, *E* number of edges ([Kuzniar *et al.*, 2010](http://www.ncbi.nlm.nih.gov/pubmed/20679333)). While *netindex* pre-processes the input graph into an efficient binary format, *netclust* performs the actual cluster detection in the graph by taking a single user-defined cutoff value (similarity or distance).

Compile and run

---------------

```

cd src

make        # default make 'all' binaries or separately 'netindex' | 'netclust'

make clean  # remove binaries

```

```

netindex ../examples/graph.tsv          # index the input (weighted) graph

netclust ../examples/graph.tsv O1 S 500 # detect clusters in the graph

```