Database schema

Introduction

cmonkey-python writes the results of its computation to an SQLite database. This choice was made, because SQLite is a free, open source and portable data store which is available on many systems and has programming interfaces to a large number of programming languages. Another important aspect is that the entire database is stored in a single file, which can be easily copied, archived and analyzed. In this section the database structure and its function is explained in further detail.

Tables

run_infos

  start_time     timestamp
  finish_time    timestamp
  num_iterations int
  last_iteration int
  organism       text
  species        text
  num_rows       int
  num_columns    int
  num_clusters   int

This table represents the current information about a cmonkey run and only stores a single entry that is continuously updated until a run is finished.

row_names, column_names

order_num int
name      text

These two tables are structurally identical. They reflect the structure of the input gene expression matrix, to preserve the order of the rows and columns, their order is stored as well.

row_members, column_members

iteration int
cluster   int
order_num int

These tables contain the row and column members for each iteration and cluster. The element order_num references an order_num in its respective row_names/column_names table.

cluster_residuals

iteration int
cluster   int
residual  decimal

Stores the residual values for each iteration and cluster.