Dataset Information

Deep learning of pharmacogenomics resources: moving towards precision oncology.

ABSTRACT: The recent accumulation of cancer genomic data provides an opportunity to understand how a tumor's genomic characteristics can affect its responses to drugs. This field, called pharmacogenomics, is a key area in the development of precision oncology. Deep learning (DL) methodology has emerged as a powerful technique to characterize and learn from rapidly accumulating pharmacogenomics data. We introduce the fundamentals and typical model architectures of DL. We review the use of DL in classification of cancers and cancer subtypes (diagnosis and treatment stratification of patients), prediction of drug response and drug synergy for individual tumors (treatment prioritization for a patient), drug repositioning and discovery and the study of mechanism/mode of action of treatments. For each topic, we summarize current genomics and pharmacogenomics data resources such as pan-cancer genomics data for cancer cell lines (CCLs) and tumors, and systematic pharmacologic screens of CCLs. By revisiting the published literature, including our in-house analyses, we demonstrate the unprecedented capability of DL enabled by rapid accumulation of data resources to decipher complex drug response patterns, thus potentially improving cancer medicine. Overall, this review provides an in-depth summary of state-of-the-art DL methods and up-to-date pharmacogenomics resources and future opportunities and challenges to realize the goal of precision oncology.

SUBMITTER: Chiu YC

PROVIDER: S-EPMC7711267 | biostudies-literature | 2020 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Deep learning of pharmacogenomics resources: moving towards precision oncology.

Chiu Yu-Chiao YC Chen Hung-I Harry HH Gorthi Aparna A Mostavi Milad M Zheng Siyuan S Huang Yufei Y Chen Yidong Y

Briefings in bioinformatics 20201201 6

The recent accumulation of cancer genomic data provides an opportunity to understand how a tumor's genomic characteristics can affect its responses to drugs. This field, called pharmacogenomics, is a key area in the development of precision oncology. Deep learning (DL) methodology has emerged as a powerful technique to characterize and learn from rapidly accumulating pharmacogenomics data. We introduce the fundamentals and typical model architectures of DL. We review the use of DL in classificat ...[more]

PMID: 31813953

Similar Datasets

Project description:The incidence of new cancer cases is expected to increase significantly in the future, posing a worldwide problem. In this regard, precision oncology and its diagnostic tools are essential for developing personalized cancer treatments. Nowadays, it is almost impossible to separate technology and digitization from medicine and health sciences, and digital pathology (DP) is emerging as one of the most influential technologies in the transition towards the 4P of new medicine (preventive, participatory, personalized and predictive). DP is a particularly key strategy to study the interactions of tumor cells and the tumor microenvironment (TME), which play a crucial role in tumor initiation, progression and metastasis. The purpose of this study was to integrate data on the digital patterns of reticulin fiber scaffolding and the immune cell infiltrate, transcriptomic and epigenetic profiles in aggressive uterine adenocarcinoma (uADC), uterine leiomyosarcoma (uLMS) and their respective lung metastases, with the aim of obtaining key TME biomarkers that can help improve metastatic prediction and shed light on potential therapeutic targets. Automatized algorithms were used to analyze reticulin fiber architecture and immune infiltration in colocalized regions of interest (ROIs) of 133 invasive tumor front (ITF), 89 tumor niches and 70 target tissues in a total of six paired samples of uADC and nine of uLMS. Microdissected tissue from the ITF was employed for transcriptomic and epigenetic studies in primary and metastatic tumors. Reticulin fiber scaffolding was characterized by a large and loose reticular fiber network in uADC, while dense bundles were found in uLMS. Notably, more similarities between reticulin fibers were observed in paired uLMS then paired uADCs. Transcriptomic and multiplex immunofluorescence-based immune profiling showed a higher abundance of T and B cells in primary tumor and in metastatic uADC than uLMS. Moreover, the epigenetic signature of paired samples in uADCs showed more differences than paired samples in uLMS. Some epigenetic variation was also found between the ITF of metastatic uADC and uLMS. Altogether, our data suggest a correlation between morphological and molecular changes at the ITF and the degree of aggressiveness. The use of DP tools for characterizing reticulin scaffolding and immune cell infiltration at the ITF in paired samples together with information provided by omics analyses in a large cohort will hopefully help validate novel biomarkers of tumor aggressiveness, develop new drugs and improve patient quality of life in a much more efficient way.

Dataset Information

Deep learning of pharmacogenomics resources: moving towards precision oncology.

Publications

Deep learning of pharmacogenomics resources: moving towards precision oncology.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets