Input Format

Locus List

You just need to specify in this field the loci. Please note that you need to use "rs" identifiers (i.e. "rs76976223 rs9443445 ..."). Please note that you need to specify at least 3 loci. Please note that you can submit maximum 100 loci.

Maximal Locus Size

You need also specify the maximal distance between an SNP and a gene on the chromosome to be considered for analysis. The default value is 200 kilobases which means that any gene located within 250 kilobases from SNP (both directions) will be considered for prioritization. Locus Spider accounts for the distance between an SNP and a gene.

Treating closed loci

In case you submit loci which are very close to each other (< 200 kilobases) they would be merged and treated as one locus.

Reference networks

In principle, Locus Spider can employ any type of information which can be used to compute the distance matrix between genes residing in genomic regions. At the moment, Locus Spider uses networks derived from public pathway databases. In the future we plan to extend the number of reference networks.


Integral_PPI_Reactome_KEGG is an integral gene network derived by merging three public databases: IntAct, Reactome and KEGG. The resulting network covers in about 10600 genes with approximately 30000 gene pairwise edges.


This is protein-protein interaction network derived from IntAct database. Only interactions reported by the papers dated 2006 and earlier are included This network is intended to be used for the benchmark purposes only.

Output Help

Top Model

The Top Model reported by Locus spider is the one which

  • is the most significant model
  • covers the maximal number of input loci
Please note that it could be several optimal models (covering different number of loci) with the same significance as the top model. As already been mentioned the top model is the one which covers the maximal number of input loci.

Browse models by the number of covered loci

By clicking this link you can view other models for a different number of covered loci. Please note that optimal models covering different numbers of loci could be notably different, i.e. they could prioritize different genes for the same locus. The Top Model is reported first on the main page. However, Locus spider infers optimal models for each l = 2, .. , n of covered input loci (where n is the number of input loci).

Browse models by input SNP

By clicking this link you can view optimal models (covering different number of input loci) which are grouped by input loci. All optimal models which cover a given locus are reported together.

Locus spider report

Locus Spider provides the same structured report for any inferred model.

Model Details

Statistical details of the model are reported first: the number of covered loci, the number (minimal) of missing genes required and model p-value (estimated by Monte Carlo simulations)

Model Online Visualization

You can view gene network model online if have Sun's Java engine installed on your machine (Version 1.5 or better). As well as your browser must be able to run Java applets (Turn Of pop-up Blocker).

Links to download graphical models

If the model exceeds 20 nodes we strongly recommend you to download networks and use Meduza program. You can create high quality figures by using this program.

Table1: prioritized genes

The table reports for each covered locus the gene being prioritized.

Table2: Missing genes (all)

The table reports all missing genes used by the model. Please note that on graphical output (Model Visualization) not all possible missing genes are shown. In many cases there are many alternative way (with the same length) to connect two loci via missing genes. This table provides all possible options.