Pervasive conditional selection of driver mutations and modular epistasis networks in cancer

Iranzo, Jaime; Gruenhagen, George; Calle-Espinosa, Jorge; Koonin, Eugene V.

doi:10.1016/j.celrep.2022.111272

Cited by 8 publications

(11 citation statements)

References 78 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This suggests that early adaptive mutations can strongly affect the direction of the adaptive process and limit the accessible adaptive paths ( SI Appendix , Fig. S1 ) ( 8 ).…”

mentioning

confidence: 99%

Fully accessible fitness landscape of oncogene-negative lung adenocarcinoma

Yousefi,

Andrejka,

Szamecz

et al. 2023

Proc. Natl. Acad. Sci. U.S.A.

View full text Add to dashboard Cite

Cancer genomes are almost invariably complex with genomic alterations cooperating during each step of carcinogenesis. In cancers that lack a single dominant oncogene mutation, cooperation between the inactivation of multiple tumor suppressor genes can drive tumor initiation and growth. Here, we shed light on how the sequential acquisition of genomic alterations generates oncogene-negative lung tumors. We couple tumor barcoding with combinatorial and multiplexed somatic genome editing to characterize the fitness landscapes of three tumor suppressor genes NF1, RASA1, and PTEN, the inactivation of which jointly drives oncogene-negative lung adenocarcinoma initiation and growth. The fitness landscape was surprisingly accessible, with each additional mutation leading to growth advantage. Furthermore, the fitness landscapes remained fully accessible across backgrounds with the inactivation of additional tumor suppressor genes. These results suggest that while predicting cancer evolution will be challenging, acquiring the multiple alterations that drive the growth of oncogene-negative tumors can be facilitated by the lack of constraints on mutational order.

show abstract

“…This suggests that early adaptive mutations can strongly affect the direction of the adaptive process and limit the accessible adaptive paths ( SI Appendix , Fig. S1 ) ( 8 ).…”

mentioning

confidence: 99%

Fully accessible fitness landscape of oncogene-negative lung adenocarcinoma

Yousefi,

Andrejka,

Szamecz

et al. 2023

Proc. Natl. Acad. Sci. U.S.A.

View full text Add to dashboard Cite

show abstract

“…To illustrate how this works in practice, we use data from a recent study that investigated epistasis between driver mutations in cancer. 1 The key idea is that epistatic interactions can be detected by looking for conditional selection on driver mutations. More precisely, epistasis implies that the effect of driver mutations in one gene depends on the presence or absence of driver mutations in another gene.…”

Section: Step-by-step Methods Detailsmentioning

confidence: 99%

“… https://gdc.cancer.gov/about-data/publications/mc3-2017 ; file mc3.v0.2.8.PUBLIC.maf.gz Example dataset - List of 441 previously reported cancer genes Iranzo et al. 1 https://www.cell.com/cms/10.1016/j.celrep.2022.111272/attachment/844802ad-e5bc-484a-b801-5906d231c25a/mmc2.xlsx Software and algorithms Coselens R package Iranzo et al. 1 https://github.com/ggruenhagen3/coselens ; RRID: SCR_022578 ; https://doi.org/10.5281/zenodo.7343763 dndscv R package Martincorena et al.…”

Section: Key Resources Tablementioning

confidence: 99%

“… 1 https://www.cell.com/cms/10.1016/j.celrep.2022.111272/attachment/844802ad-e5bc-484a-b801-5906d231c25a/mmc2.xlsx Software and algorithms Coselens R package Iranzo et al. 1 https://github.com/ggruenhagen3/coselens ; RRID: SCR_022578 ; https://doi.org/10.5281/zenodo.7343763 dndscv R package Martincorena et al. 2 https://www.sanger.ac.uk/tool/dndscv/ ; RRID: SCR_017093 CrossMap Zhao et al.…”

Section: Key Resources Tablementioning

confidence: 99%

“…The last two columns (pval and qval) report the statistical significance and false discovery rate of differential selection, obtained through a likelihood ratio test as described in Iranzo et al . 1 The false discovery rate is obtained by applying the Benjamini-Hochberg correction for multiple testing to all the genes shown in the table.…”

Section: Expected Outcomesmentioning

confidence: 99%

See 2 more Smart Citations

Protocol for comparing gene-level selection on coding mutations between two groups of samples with Coselens

et al. 2023

Self Cite

View full text Add to dashboard Cite

Overcoming Observation Bias for Cancer Progression Modeling

Schill,

Klever,

Lösch

et al. 2024

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Cancers evolve by accumulating genetic alterations, such as mutations and copy number changes. The chronological order of these events is important for understanding the disease, but not directly observable from cross-sectional genomic data. Cancer progression models (CPMs), such as Mutual Hazard Networks (MHNs), reconstruct the progression dynamics of tumors by learning a network of causal interactions between genetic events from their co-occurrence patterns. However, current CPMs fail to include effects of genetic events on the observation of the tumor itself and assume that observation occurs independently of all genetic events. Since a dataset contains by definition only tumors at their moment of observation, neglecting any causal effects on this event leads to the "conditioning on a collider" bias: Events that make the tumor more likely to be observed appear anti-correlated, which results in spurious suppressive effects or masks promoting effects among genetic events. Here, we extend MHNs by modeling effects from genetic progression events on the observation event, thereby correcting for the collider bias. We derive an efficient tensor formula for the likelihood function and learn two models on somatic mutation datasets from the MSK-IMPACT study. In colon adenocarcinoma, we find a strong effect on observation by mutations in TP53, and in lung adenocarcinoma by mutations in EGFR. Compared to classical MHNs, this explains away many spurious suppressive interactions and uncovers several promoting effects.The data, code, and results are available at https://github.com/cbg-ethz/ObservationMHN.

show abstract

Pervasive conditional selection of driver mutations and modular epistasis networks in cancer

Cited by 8 publications

References 78 publications

Fully accessible fitness landscape of oncogene-negative lung adenocarcinoma

Fully accessible fitness landscape of oncogene-negative lung adenocarcinoma

Protocol for comparing gene-level selection on coding mutations between two groups of samples with Coselens

Overcoming Observation Bias for Cancer Progression Modeling

Contact Info

Product

Resources

About