Global computational alignment of tumor and cell line transcriptional profiles

Warren, Allison; Chen, Yejia; Jones, Andrew; Shibue, Tsukasa; Hahn, William C.; Boehm, Jesse S.; Vázquez, Francisca; Tsherniak, Aviad; McFarland, James M.

doi:10.1038/s41467-020-20294-x

Cited by 108 publications

(107 citation statements)

References 83 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When systematic in vivo screens become available, the proposed model can be a stepping stone toward an accurate prediction for tumor dependencies. Furthermore, methods that align genomic profiles between CCLs and tumors ( 40 , 41 ) may help to reduce the differences between tumor and CCL domains. We expect that the incorporation of these methods will improve the translational capability of DeepDEP along with the expansion of CCLs being screened by the DepMap projects.…”

Section: Discussionmentioning

confidence: 99%

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

Chiu¹,

Zheng²,

Wang³

et al. 2021

Sci. Adv.

View full text Add to dashboard Cite

Genome-wide loss-of-function screens have revealed genes essential for cancer cell proliferation, called cancer dependencies. It remains challenging to link cancer dependencies to the molecular compositions of cancer cells or to unscreened cell lines and further to tumors. Here, we present DeepDEP, a deep learning model that predicts cancer dependencies using integrative genomic profiles. It uses a unique unsupervised pretraining that captures unlabeled tumor genomic representations to improve the learning of cancer dependencies. We demonstrated DeepDEP's improvement over conventional machine learning methods and validated the performance with three independent datasets. By systematic model interpretations, we extended the current dependency maps with functional characterizations of dependencies and a proof-of-concept in silico assay of synthetic essentiality. We applied DeepDEP to pan-cancer tumor genomics and built the first pan-cancer synthetic dependency map of 8000 tumors with clinical relevance. In summary, DeepDEP is a novel tool for investigating cancer dependency with rapidly growing genomic resources.

show abstract

Section: Discussionmentioning

confidence: 99%

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

Chiu¹,

Zheng²,

Wang³

et al. 2021

Sci. Adv.

View full text Add to dashboard Cite

show abstract

“…Despite the undeniable value of these data, molecular differences between these models and patient tumors call for caution in extending results to the human system [ 128 , 129 ]. Therefore, approaches aiming at delineating the similarities and differences between preclinical and clinical transcriptomes are required for an effective application of AI to improve the patient’s quality of life [ 130 , 131 ].…”

Section: Discussionmentioning

confidence: 99%

Artificial Intelligence in Bulk and Single-Cell RNA-Sequencing Data to Foster Precision Oncology

Giudice

Peirone

Perrone

et al. 2021

IJMS

View full text Add to dashboard Cite

Artificial intelligence, or the discipline of developing computational algorithms able to perform tasks that requires human intelligence, offers the opportunity to improve our idea and delivery of precision medicine. Here, we provide an overview of artificial intelligence approaches for the analysis of large-scale RNA-sequencing datasets in cancer. We present the major solutions to disentangle inter- and intra-tumor heterogeneity of transcriptome profiles for an effective improvement of patient management. We outline the contributions of learning algorithms to the needs of cancer genomics, from identifying rare cancer subtypes to personalizing therapeutic treatments.

show abstract

“…It remains uncertain how well cell lines reflect the biological characteristics of tumors. Systematic differences between cell lines and human cancers may be due to many factors such as culture conditions, clonal selection, and genomic instability ( 70 ). So, despite these promising results and their clinical implications, we should acknowledge the limitations, as well as directions for future research.…”

Section: Discussionmentioning

confidence: 99%

Comprehensive Analysis of E2F Family Members in Human Gastric Cancer

Yang

et al. 2021

Front. Oncol.

View full text Add to dashboard Cite

Gastric cancer (GC) is the second most common cancer and the third most frequent cause of cancer-related deaths in China. E2Fs are a family of transcription factors reported to be involved in the tumor progression of various cancer types; however, the roles of individual E2Fs are still not known exactly in tumor progression of GC. In this study, we examined the expression of E2Fs to investigate their roles in tumor progression in GC patients using multiple databases, including ONCOMINE, GEPIA2, Kaplan-Meier plotter, cBioPortal, Metascape, LinkedOmics, GeneMANIA, STRING and UCSC Xena. We also performed real-time polymerase chain reaction (RT-PCR) to validate the expression levels of individual E2Fs in several GC cell lines. Our results demonstrated that the mRNA levels of E2F1/2/3/5/8 were significantly higher both in GC tissues and cell lines. The expression levels of E2F1 and E2F4 were correlated with poor overall survival (OS), decreased post-progression survival (PPS), and decreased progression-free survival (FP) in patients with GC. However, overexpression of E2F2, E2F5, E2F7 and E2F8 is significantly associated with disease-free survival and overall survival in patients with GC. In addition, higher E2F3 and E2F6 mRNA expression was found to increase GC patients’ OS and PPS. 224 of 415 patients with STAD (54%) had gene mutations that were associated with longer disease-free survival (DFS) but not OS. Cell cycle pathway was closely associated with mRNA level of more than half of E2Fs (E2F1/2/3/7/8). There were close and complicated interactions among E2F family members. Finally, our results indicated the gene expressions of E2Fs had a positive relationship with its copy numbers. Taken together, E2F1/2/3/5/8 can serve as biomarkers for GC patients with high prognostic value for OS of GC patients or therapeutic targets for GC.

show abstract

Global computational alignment of tumor and cell line transcriptional profiles

Cited by 108 publications

References 83 publications

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

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

Artificial Intelligence in Bulk and Single-Cell RNA-Sequencing Data to Foster Precision Oncology

Comprehensive Analysis of E2F Family Members in Human Gastric Cancer

Contact Info

Product

Resources

About