L1L2Signature - Unbiased framework for -omics data analysis

Release:0.2.2 Homepage:http://slipguru.disi.unige.it/Software/L1L2Signature Repository:https://bitbucket.org/slipguru/l1l2signature

L1L2Signature is an implementation of an unbiased framework originally thought for gene expression analysis. The need of such a framework may be found in the Framework overview section illustrating the dramatic effect of a biased approach especially when the sample size is small.

The framework, here implemented, was used in many real applications and a collection of them is referenced in the Framework Applications section.

This library is composed by a set of Python scripts (described in the Quick start tutorial) and a set of useful classes and functions (described in the Public API section) that could be used to manually read and/or analyze high-throughput data extending/integrating the proposed pipeline.

L1L2Signature relies on two libraries also implemented by our research group and already released as open source libraries:

• L1L2Py: implements the gene selection core, based on elastic net regularization.
• PPlus: used to parallelize cross validation splits in an easy and effective way across a set of desktop personal computer distributed in our labs.

User documentation

• Framework overview
  • Model (gene) selection
  • Model assessment
• Framework Applications
• Quick start tutorial
  • Installation from sources
  • Input data format
  • Configuration File
  • Tau choice helper
  • Experiment runner
  • Results analysis

Public API

• Utility functions and classes (utils)
• Plotting functions (plots)

Indices and tables

• Index
• Search Page