FIREWORKS: a bottom-up approach to integrative coessentiality network analysis

Abstract

Genetic coessentiality analysis, a computational approach which identifies genes sharing a common effect on cell fitness across large-scale screening datasets, has emerged as a powerful tool to identify functional relationships between human genes. However, widespread implementation of coessentiality to study individual genes and pathways is limited by systematic biases in existing coessentiality approaches and accessibility barriers for investigators without computational expertise. We created FIREWORKS, a method and interactive tool for the construction and statistical analysis of coessentiality networks centered around gene(s) provided by the user. FIREWORKS incorporates a novel bias reduction approach to reduce false discoveries, enables restriction of coessentiality analyses to custom subsets of cell lines, and integrates multiomic and drug–gene interaction datasets to investigate and target contextual gene essentiality. We demonstrate the broad utility of FIREWORKS through case vignettes investigating gene function and specialization, indirect therapeutic targeting of “undruggable” proteins, and context-specific rewiring of genetic networks.

Sparse dictionary learning recovers pleiotropy from human cell fitness screens

“…Nineteen of the dictionary elements were highly loaded for essential genes, which incur fitness effects at the lower detection limit of our assay and group together in our analysis for that reason. These functions were labeled as ''Common Essential (Gene)'', where the Gene was chosen to be the top loaded gene, or ''Common Essential (Chr#)'', when the common essential genes shared synteny on chromosome regions (Amici et al, 2021). Finally, two dictionary elements could not be mapped to a biological process.…”

Predicting and characterizing a cancer dependency map of tumors with deep learning

“…Expanding the range of biological contexts, computational tools such as CEN-tools ( 55 ) identify dependencies that are specific to the tissue of origin and/or individual mutation state and expression level. On the basis of the concepts of differential network biology ( 56 ) and genetic coessentiality ( 57 ), an interactive tool called FIREWORKS ( 58 ) interrogates functional relationships between gene dependencies using context-specific “coessentiality networks” and identifies multiomic signatures associated with differential gene dependencies. Since our goal is to link multiomics to the landscape of dependencies regardless of their simple or complex associations, we designed a unique architecture to allow the model to learn from genes with similar functions implemented by a functional fingerprint.…”

Sparse dictionary learning recovers pleiotropy from human cell fitness screens

“…To achieve this, we exploit several properties of fitness data. First, genes with correlated fitness effects tend to have similar biological functions, a concept referred to as co-essentiality (Amici et al, 2021;Bayraktar et al, 2020;Boyle et al, 2018;Kim et al, 2019;Pan et al, 2018;Wainberg et al, 2021;Wang et al, 2017). Second, similar cell contexts will share similar rate limiting functions for cell growth.…”