FIREWORKS (Fitness Interaction Ranked nEtWORKS)

FIREWORKS is an interactive web tool developed to facilitate interrogation of biological networks using unbiased genetic screens. Gene-gene relationships can be modeled by correlating essentiality scores across hundreds of cancer cell lines. This application allows you to visualize these relationships, build networks from them and leverage multi-omics data to investigate their underlying biology.

Getting started

• Network:
  Any list of genes can be used as the foundation of a network. These genes are referred to as “source nodes”. The most coessential genes for each source node form the “primary nodes". The most coessential genes for each primary node can be added by selecting “secondary nodes”.
  
  Adding secondary nodes increases the compute time, but often results in genes organizing into functional modules. Frequently, these networks return nodes that are not connected to any other sub-networks. This can result in a cluttered looking visualization. To simplify these networks, we have provided an option to remove isolated primary and secondary nodes.

• Coessentiality Heatmap:
  In the “coessentiality heatmap” tab, you can investigate the relative gene essentiality of a gene of interest and its top coessential partners across cell lines. You must specify the context, correlation direction and rank cut-off for primary connections. Second order connections are not included.

• Multi-omics:
  What determines which cell lines are dependent on a given gene signature? The “multi-omics” tab aids in answering this question, by integrating fitness data with other functional characterizations of cancer cell lines. Currently, you can use the “RNA” tool to perform differential expression between the cell lines that are the most and least dependent on a given gene, or gene signature. The analysis depends on the provided context, p-value cutoff and quantile cut-off.
  
  IMPORTANT NOTE: Signatures containing anti-correlated genes will still group cell lines by dependence and perform differential expression between these groups. This may not provide biologically relevant results.

Contact Us

This tool is a work in progress. Please feel free to contact us with any questions, comments or other related inquiries. We will get back to you as soon as possible (david.amici@northwestern.edu).

Read the paper

FIREWORKS: a bottom-up approach to integrative coessentiality network analysis
David R Amici, Jasen M Jackson, Mihai I Truica, Roger S Smith, Byoung-Kyu Cho, Sarki A Abdulkadir, Marc L Mendillo
Life Science Alliance 2020