Gene isoforms as expression-based biomarkers predictive of drug response in vitro

Safikhani, Zhaleh; Смирнов, Петр; Thu, Kelsie L.; Silvester, Jennifer; El-Hachem, Nehmé; Quevedo, Rene; Lupien, Mathieu; Mak, Tak W.; Cescon, David W.; Haibe‐Kains, Benjamin

doi:10.1038/s41467-017-01153-8

Cited by 53 publications

(25 citation statements)

References 62 publications

(79 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, our pilot study showed that the MS-based proteomics can significantly improve the drug response predictions, but only after filtering out most of the protein measurements (Ali et al 2018 ). Similarly, a recent transcript-level machine learning work demonstrates how the RNA-seq technology offers additional predictive signal, when compared to gene-level expression or mutation information (Safikhani et al 2017 ). Therefore, we argue that we will need improvements both in the computational methods and in the experimental assays in order to convincingly show the added value of “big data” for drug response prediction.…”

Section: Discussionmentioning

confidence: 99%

Machine learning and feature selection for drug response prediction in precision oncology applications

2018

View full text Add to dashboard Cite

In-depth modeling of the complex interplay among multiple omics data measured from cancer cell lines or patient tumors is providing new opportunities toward identification of tailored therapies for individual cancer patients. Supervised machine learning algorithms are increasingly being applied to the omics profiles as they enable integrative analyses among the high-dimensional data sets, as well as personalized predictions of therapy responses using multi-omics panels of response-predictive biomarkers identified through feature selection and cross-validation. However, technical variability and frequent missingness in input "big data" require the application of dedicated data preprocessing pipelines that often lead to some loss of information and compressed view of the biological signal. We describe here the state-of-the-art machine learning methods for anti-cancer drug response modeling and prediction and give our perspective on further opportunities to make better use of high-dimensional multi-omics profiles along with knowledge about cancer pathways targeted by anti-cancer compounds when predicting their phenotypic responses.

show abstract

Section: Discussionmentioning

confidence: 99%

Machine learning and feature selection for drug response prediction in precision oncology applications

2018

View full text Add to dashboard Cite

show abstract

“…Aneuploidy Cluster Membership was obtained from the published supplemental section [38]. The CCLE transcript RNA-seq dataset was used for comparison of FOXM1 isoform mRNA expression in human cancer cell lines [69].…”

Section: Methodsmentioning

confidence: 99%

Pan-Cancer Analyses Reveal Genomic Features of FOXM1 Overexpression in Cancer

Barger

Branick

Chee

et al. 2019

Cancers

156

153

View full text Add to dashboard Cite

FOXM1 is frequently overexpressed in cancer, but this has not been studied in a comprehensive manner. We utilized genotype-tissue expression (GTEx) normal and The Cancer Genome Atlas (TCGA) tumor data to define FOXM1 expression, including its isoforms, and to determine the genetic alterations that promote FOXM1 expression in cancer. Additionally, we used human fallopian tube epithelial (FTE) cells to dissect the role of Retinoblastoma (Rb)-E2F and Cyclin E1 in FOXM1 regulation, and a novel human embryonic kidney cell (HEK293T) CRISPR FOXM1 knockout model to define isoform-specific transcriptional programs. FOXM1 expression, at the mRNA and protein level, was significantly elevated in tumors with FOXM1 amplification, p53 inactivation, and Rb-E2F deregulation. FOXM1 expression was remarkably high in testicular germ cell tumors (TGCT), high-grade serous ovarian cancer (HGSC), and basal breast cancer (BBC). FOXM1 expression in cancer was associated with genomic instability, as measured using aneuploidy signatures. FTE models confirmed a role for Rb-E2F signaling in FOXM1 regulation and in particular identified Cyclin E1 as a novel inducer of FOXM1 expression. Among the three FOXM1 isoforms, FOXM1c showed the highest expression in normal and tumor tissues and cancer cell lines. The CRISPR knockout model demonstrated that FOXM1b and FOXM1c are transcriptionally active, while FOXM1a is not. Finally, we were unable to confirm the existence of a FOXM1 auto-regulatory loop. This study provides significant and novel information regarding the frequency, causes, and consequences of elevated FOXM1 expression in human cancer.

show abstract

“…Future work may also be motivated by the fact that the CCLE RNA-Seq dataset could allow the analysis of transcript-level (gene isoform) data for predicting drug response. Such information has been recently shown to be a useful source of features for drug sensitivity prediction 49 . Moreover, the investigation of the biological role of hubs in gene isoform networks may open new directions for drug sensitivity research and other applications.…”

Section: Discussionmentioning

confidence: 99%

Hub genes in a pan-cancer co-expression network show potential for predicting drug responses

et al. 2019

View full text Add to dashboard Cite

Background: The topological analysis of networks extracted from different types of “omics” data is a useful strategy for characterizing biologically meaningful properties of the complex systems underlying these networks. In particular, the biological significance of highly connected genes in diverse molecular networks has been previously determined using data from several model organisms and phenotypes. Despite such insights, the predictive potential of candidate hubs in gene co-expression networks in the specific context of cancer-related drug experiments remains to be deeply investigated. The examination of such associations may offer opportunities for the accurate prediction of anticancer drug responses. Methods: Here, we address this problem by: a) analyzing a co-expression network obtained from thousands of cancer cell lines, b) detecting significant network hubs, and c) assessing their capacity to predict drug sensitivity using data from thousands of drug experiments. We investigated the prediction capability of those genes using a multiple linear regression model, independent datasets, comparisons with other models and our own in vitro experiments. Results: These analyses led to the identification of 47 hub genes, which are implicated in a diverse range of cancer-relevant processes and pathways. Overall, encouraging agreements between predicted and observed drug sensitivities were observed in public datasets, as well as in our in vitro validations for four glioblastoma cell lines and four drugs. To facilitate further research, we share our hub-based drug sensitivity prediction model as an online tool. Conclusions: Our research shows that co-expression network hubs are biologically interesting and exhibit potential for predicting drug responses in vitro. These findings motivate further investigations about the relevance and application of our unbiased discovery approach in pre-clinical, translationally-oriented research.

show abstract

Gene isoforms as expression-based biomarkers predictive of drug response in vitro

Cited by 53 publications

References 62 publications

Machine learning and feature selection for drug response prediction in precision oncology applications

Machine learning and feature selection for drug response prediction in precision oncology applications

Pan-Cancer Analyses Reveal Genomic Features of FOXM1 Overexpression in Cancer

Hub genes in a pan-cancer co-expression network show potential for predicting drug responses

Contact Info

Product

Resources

About