About this supplement

Installation

Start with VizRank in Orange

VizRank's graphical interface

VizRank method details

Supported visualization methods

Comparison with other methods



S. cerevisiae metabolic example

S. cerevisiae cell cycle example

Spotlight

Try VizRank online - You can now experiment with VizRank online. Find interesting data projections of your own data sets.
[This now works again]


FRI > Biolab > Supplements > VizRank

Orange Data Visualization

Step 1 of 3: Data File Input

Orange Data Visualization is a server-based application that searches for interesting projections of a given data set. Projections are assessed by inducing a k-nearest neighbor classifier on the projected data and evaluating its classification accuracy. Data set files that can be accepted are tab-delimited text files where each column represents an attribute and each row represents a data instance. The file must also have a header with attribute name and data type. For details on suitable formats see Loading The Data section of the Orange manual.

Purpose of Orange Data Visualization is to quickly find some interesting data projections. For serious use we advise you to install Orange. This way you will get a wide range of data mining and visualization components (called widgets), including VizRank.

To proceed, upload a data file (press Browse button):



or pick any data set from bottom list:

Data sets from functional genomics:
   budding yeast Saccharomyces cerevisiae - three functional groups from budding yeast Saccharomyces cerevisiae data set (186 instances,  79 continuous attributes)
   yeast cell-cycle - expression of genes from two cell cycle phases, G1 and S/G2 (421 instances,  77 continuous attributes)

Data sets from cancer research:
   Leukemia data set - 48 acute lymphoblastic leukemia (ALL) samples and 25 acute myeloid leukemia (AML) samples (7074 attributes (genes), 73 instances (samples))
   SRBCT - The small round blue cell tumors (SRBCT) data set consists of four types of tumors in childhood, including Ewing's sarcoma (EWS), rhabdomyosarcoma (RB), neuroblastoma (NB) and Burkitt's lymphoma (BL) (2,308 attributes (genes), 83 instances (samples)

Data sets from machine learning repository:
   Wine (178 instances, 13 continuous attributes)
   Imports-85 (205 instances, 11 discrete attributes,  14 continuous attributes, class is discretized)
   Voting (436 instances, 16 discrete attributes)