scholarly journals CancerSiamese: one-shot learning for primary and metastatic tumor classification

2020 ◽  
Author(s):  
Milad Mostav ◽  
Yu-Chiao Chiu ◽  
Yidong Chen ◽  
Yufei Huang
Keyword(s):  
Gene Expression ◽  
Expression Profiles ◽  
Metastatic Cancer ◽  
Single Gene ◽  
Marker Gene ◽  
Metastatic Tumor ◽  
Saliency Map ◽  
Similarity Function ◽  
Cancer Type ◽  
Cancer Types

AbstractWe consider cancer classification based on one single gene expression profile. We proposed CancerSiamese, a new one-shot learning model, to predict the cancer type of a query primary or metastatic tumor sample based on a support set that contains only one known sample for each cancer type. CancerSiamese receives pairs of gene expression profiles and learns a representation of similar or dissimilar cancer types through two parallel Convolutional Neural Networks joined by a similarity function. We trained CancerSiamese for both primary and metastatic cancer type predictions using samples from TCGA and MET500. Test results for different N-way predictions yielded an average accuracy improvement of 8% and 4% over the benchmark 1-Nearest Neighbor (1-NN) classifier for primary and metastatic tumors, respectively. Moreover, we applied the guided gradient saliency map and feature selection to CancerSiamese to identify and analyze the marker-gene candidates for primary and metastatic cancers. Our work demonstrated, for the first time, the feasibility of applying one-shot learning for expression-based cancer type prediction when gene expression data of cancer types are limited and could inspire new and ingenious applications of one-shot and few-shot learning solutions for improving cancer diagnosis, treatment planning, and our understanding of cancer.

scholarly journals CancerSiamese: one-shot learning for predicting primary and metastatic tumor types unseen during model training

2021 ◽  
Vol 22 (1) ◽  
Author(s):  
Milad Mostavi ◽  
Yu-Chiao Chiu ◽  
Yidong Chen ◽  
Yufei Huang
Keyword(s):  
Nearest Neighbor ◽  
Expression Profiles ◽  
Marker Gene ◽  
Saliency Map ◽  
The Cancer Genome Atlas ◽  
Marker Genes ◽  
Similarity Function ◽  
Cancer Type ◽  
Metastatic Tumors ◽  
Cancer Types

Abstract Background The state-of-the-art deep learning based cancer type prediction can only predict cancer types whose samples are available during the training where the sample size is commonly large. In this paper, we consider how to utilize the existing training samples to predict cancer types unseen during the training. We hypothesize the existence of a set of type-agnostic expression representations that define the similarity/dissimilarity between samples of the same/different types and propose a novel one-shot learning model called CancerSiamese to learn this common representation. CancerSiamese accepts a pair of query and support samples (gene expression profiles) and learns the representation of similar or dissimilar cancer types through two parallel convolutional neural networks joined by a similarity function. Results We trained CancerSiamese for cancer type prediction for primary and metastatic tumors using samples from the Cancer Genome Atlas (TCGA) and MET500. Network transfer learning was utilized to facilitate the training of the CancerSiamese models. CancerSiamese was tested for different N-way predictions and yielded an average accuracy improvement of 8% and 4% over the benchmark 1-Nearest Neighbor (1-NN) classifier for primary and metastatic tumors, respectively. Moreover, we applied the guided gradient saliency map and feature selection to CancerSiamese to examine 100 and 200 top marker-gene candidates for the prediction of primary and metastatic cancers, respectively. Functional analysis of these marker genes revealed several cancer related functions between primary and metastatic tumors. Conclusion This work demonstrated, for the first time, the feasibility of predicting unseen cancer types whose samples are limited. Thus, it could inspire new and ingenious applications of one-shot and few-shot learning solutions for improving cancer diagnosis, prognostic, and our understanding of cancer.

scholarly journals Analyzing cancer gene expression data through the lens of normal tissue-specificity

2021 ◽  
Author(s):  
H. Robert Frost
Keyword(s):  
Gene Expression ◽  
Normal Tissue ◽  
Tissue Specificity ◽  
Cancer Genomics ◽  
Expression Profiles ◽  
Genetic Alterations ◽  
Cancer Type ◽  
Tissue Specific ◽  
Normal Tissues ◽  
Cancer Types

AbstractThe genetic alterations that underlie cancer development are highly tissue-specific with the majority of driving alterations occurring in only a few cancer types and with alterations common to multiple cancer types often showing a tissue-specific functional impact. This tissue-specificity means that the biology of normal tissues carries important information regarding the pathophysiology of the associated cancers, information that can be leveraged to improve the power and accuracy of cancer genomic analyses. Research exploring the use of normal tissue data for the analysis of cancer genomics has primarily focused on the paired analysis of tumor and adjacent normal samples. Efforts to leverage the general characteristics of normal tissue for cancer analysis has received less attention with most investigations focusing on understanding the tissue-specific factors that lead to individual genomic alterations or dysregulated pathways within a single cancer type. To address this gap and support scenarios where adjacent normal tissue samples are not available, we explored the genome-wide association between the transcriptomes of 21 solid human cancers and their associated normal tissues as profiled in healthy individuals. While the average gene expression profiles of normal and cancerous tissue may appear distinct, with normal tissues more similar to other normal tissues than to the associated cancer types, when transformed into relative expression values, i.e., the ratio of expression in one tissue or cancer relative to the mean in other tissues or cancers, the close association between gene activity in normal tissues and related cancers is revealed. As we demonstrate through an analysis of tumor data from The Cancer Genome Atlas and normal tissue data from the Human Protein Atlas, this association between tissue-specific and cancer-specific expression values can be leveraged to improve the prognostic modeling of cancer, the comparative analysis of different cancer types, and the analysis of cancer and normal tissue pairs.

scholarly journals Analyzing cancer gene expression data through the lens of normal tissue-specificity

2021 ◽  
Vol 17 (6) ◽  
pp. e1009085
Author(s):  
H. Robert Frost
Keyword(s):  
Gene Expression ◽  
Normal Tissue ◽  
Tissue Specificity ◽  
Cancer Genomics ◽  
Expression Profiles ◽  
Genetic Alterations ◽  
Cancer Type ◽  
Tissue Specific ◽  
Normal Tissues ◽  
Cancer Types

The genetic alterations that underlie cancer development are highly tissue-specific with the majority of driving alterations occurring in only a few cancer types and with alterations common to multiple cancer types often showing a tissue-specific functional impact. This tissue-specificity means that the biology of normal tissues carries important information regarding the pathophysiology of the associated cancers, information that can be leveraged to improve the power and accuracy of cancer genomic analyses. Research exploring the use of normal tissue data for the analysis of cancer genomics has primarily focused on the paired analysis of tumor and adjacent normal samples. Efforts to leverage the general characteristics of normal tissue for cancer analysis has received less attention with most investigations focusing on understanding the tissue-specific factors that lead to individual genomic alterations or dysregulated pathways within a single cancer type. To address this gap and support scenarios where adjacent normal tissue samples are not available, we explored the genome-wide association between the transcriptomes of 21 solid human cancers and their associated normal tissues as profiled in healthy individuals. While the average gene expression profiles of normal and cancerous tissue may appear distinct, with normal tissues more similar to other normal tissues than to the associated cancer types, when transformed into relative expression values, i.e., the ratio of expression in one tissue or cancer relative to the mean in other tissues or cancers, the close association between gene activity in normal tissues and related cancers is revealed. As we demonstrate through an analysis of tumor data from The Cancer Genome Atlas and normal tissue data from the Human Protein Atlas, this association between tissue-specific and cancer-specific expression values can be leveraged to improve the prognostic modeling of cancer, the comparative analysis of different cancer types, and the analysis of cancer and normal tissue pairs.

Gene expression data analysis for characterizing shared and type specific mechanisms of HCC and B-CLL

2018 ◽  
Vol 44 (1) ◽  
pp. 86-97
Author(s):  
Ceren Sucularli ◽  
Ugur Toprak ◽  
Melda Arslantas
Keyword(s):  
Gene Expression ◽  
Cell Cycle ◽  
B Cell ◽  
Expression Profiles ◽  
Gene Expression Profiles ◽  
B Cell Lymphoma ◽  
Specific Gene ◽  
Cancer Type ◽  
Expression Data ◽  
Cancer Types

Abstract Background Comparing gene expression profiles using gene expression datasets of different types of tumors is frequently used to identify molecular mechanisms of cancer. This study aimed to find shared and type specific gene expression profiles of hepatocellular carcinoma (HCC) and B-cell chronic lymphocytic leukemia (B-CLL). Material and methods Gene expression microarrays for HCC and B-CLL and RNA-sequencing expression data for liver HCC and lymphoid neoplasm diffuse large B-cell lymphoma (DLBC) were analyzed and differentially expressed probe sets or genes for each cancer type were detected. Probe sets and genes that were shared or specifically expressed in both cancer types were identified. Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway and Gene Ontology (GO) terms for Biological Process (BP) annotations were performed. Results According to our analysis shared upregulated genes were mainly annotated in cell cycle processes. Some of the genes that changed only in HCC were annotated in cell cycle and metabolic processes, and some of the genes, altered only in B-CLL, were annotated in immune response and hemopoiesis. Conclusion These results contribute to cancer research that aim to find the conserved gene expression profiles in different cancer types and widen the knowledge of HCC and B-CLL specific mechanisms.

scholarly journals Pan-Cancer Transcriptional Models Predicting Chemosensitivity in Human Tumors

2021 ◽  
Vol 20 ◽  
pp. 117693512110024
Author(s):  
Jason D Wells ◽  
Jacqueline R Griffin ◽  
Todd W Miller
Keyword(s):  
Gene Expression ◽  
Expression Profiles ◽  
Human Tumors ◽  
Molecular Characteristics ◽  
Survival Times ◽  
Chemotherapy Drugs ◽  
Testing Dataset ◽  
Cancer Models ◽  
Cancer Types ◽  
Pan Cancer

Motivation: Despite increasing understanding of the molecular characteristics of cancer, chemotherapy success rates remain low for many cancer types. Studies have attempted to identify patient and tumor characteristics that predict sensitivity or resistance to different types of conventional chemotherapies, yet a concise model that predicts chemosensitivity based on gene expression profiles across cancer types remains to be formulated. We attempted to generate pan-cancer models predictive of chemosensitivity and chemoresistance. Such models may increase the likelihood of identifying the type of chemotherapy most likely to be effective for a given patient based on the overall gene expression of their tumor. Results: Gene expression and drug sensitivity data from solid tumor cell lines were used to build predictive models for 11 individual chemotherapy drugs. Models were validated using datasets from solid tumors from patients. For all drug models, accuracy ranged from 0.81 to 0.93 when applied to all relevant cancer types in the testing dataset. When considering how well the models predicted chemosensitivity or chemoresistance within individual cancer types in the testing dataset, accuracy was as high as 0.98. Cell line–derived pan-cancer models were able to statistically significantly predict sensitivity in human tumors in some instances; for example, a pan-cancer model predicting sensitivity in patients with bladder cancer treated with cisplatin was able to significantly segregate sensitive and resistant patients based on recurrence-free survival times ( P = .048) and in patients with pancreatic cancer treated with gemcitabine ( P = .038). These models can predict chemosensitivity and chemoresistance across cancer types with clinically useful levels of accuracy.

scholarly journals Estimating the predictive power of silent mutations on cancer classification and prognosis

2021 ◽  
Vol 6 (1) ◽  
Author(s):  
Tal Gutman ◽  
Guy Goren ◽  
Omri Efroni ◽  
Tamir Tuller
Keyword(s):  
Gene Expression ◽  
Predictive Power ◽  
Predictive Ability ◽  
Cancer Classification ◽  
Silent Mutation ◽  
Cancer Type ◽  
Coding Region ◽  
Probability Of Survival ◽  
Diagnosis And Prognosis ◽  
Cancer Types

AbstractIn recent years it has been shown that silent mutations, in and out of the coding region, can affect gene expression and may be related to tumorigenesis and cancer cell fitness. However, the predictive ability of these mutations for cancer type diagnosis and prognosis has not been evaluated yet. In the current study, based on the analysis of 9,915 cancer genomes and approximately three million mutations, we provide a comprehensive quantitative evaluation of the predictive power of various types of silent and non-silent mutations over cancer classification and prognosis. The results indicate that silent-mutation models outperform the equivalent null models in classifying all examined cancer types and in estimating the probability of survival 10 years after the initial diagnosis. Additionally, combining both non-silent and silent mutations achieved the best classification results for 68% of the cancer types and the best survival estimation results for up to nine years after the diagnosis. Thus, silent mutations hold considerable predictive power over both cancer classification and prognosis, most likely due to their effect on gene expression. It is highly advised that silent mutations are integrated in cancer research in order to unravel the full genomic landscape of cancer and its ramifications on cancer fitness.

High-throughput single-cell RNA-seq data imputation and characterization with surrogate-assisted automated deep learning

2021 ◽  
Author(s):  
Xiangtao Li ◽  
Shaochuan Li ◽  
Lei Huang ◽  
Shixiong Zhang ◽  
Ka-chun Wong
Keyword(s):  
Gene Expression ◽  
Neural Networks ◽  
Single Cell ◽  
Deep Neural Networks ◽  
Expression Profiles ◽  
Marker Gene ◽  
Gene Expression Profiles ◽  
Underlying Mechanisms ◽  
Cell Data ◽  
Gene Expression Levels

Abstract Single-cell RNA sequencing (scRNA-seq) technologies have been heavily developed to probe gene expression profiles at single-cell resolution. Deep imputation methods have been proposed to address the related computational challenges (e.g. the gene sparsity in single-cell data). In particular, the neural architectures of those deep imputation models have been proven to be critical for performance. However, deep imputation architectures are difficult to design and tune for those without rich knowledge of deep neural networks and scRNA-seq. Therefore, Surrogate-assisted Evolutionary Deep Imputation Model (SEDIM) is proposed to automatically design the architectures of deep neural networks for imputing gene expression levels in scRNA-seq data without any manual tuning. Moreover, the proposed SEDIM constructs an offline surrogate model, which can accelerate the computational efficiency of the architectural search. Comprehensive studies show that SEDIM significantly improves the imputation and clustering performance compared with other benchmark methods. In addition, we also extensively explore the performance of SEDIM in other contexts and platforms including mass cytometry and metabolic profiling in a comprehensive manner. Marker gene detection, gene ontology enrichment and pathological analysis are conducted to provide novel insights into cell-type identification and the underlying mechanisms. The source code is available at https://github.com/li-shaochuan/SEDIM.

Expressed genome molecular signatures of heart failure

2005 ◽  
Vol 43 (5) ◽  
Author(s):  
Choong Chin Liew
Keyword(s):  
Gene Expression ◽  
Heart Failure ◽  
Blood Cells ◽  
Complex Disease ◽  
Expression Profiles ◽  
Single Gene ◽  
Molecular Signatures ◽  
Microarray Gene Expression ◽  
Genomic Technologies ◽  
Gene Expression Studies

AbstractTraditional gene expression studies typically focus on one or a few genes of interest. An important limitation of single-gene studies is that they present a portrait of disease that is essentially static. However, disease is a dynamic process, driven by a combination of genetic, epigenetic and environmental factors. Recently, genomic technologies have permitted better characterization of the dynamic aspect of disease progression. Genome-wide expression profiles of cardiovascular diseases, heart failure in particular, using microarrays have been published and are providing new insights into this complex disease. Tissue biopsies required for traditional microarray studies, however, are often invasive and not readily available. By contrast, blood samples are relatively non-invasive and are readily available. In a number of recent studies, blood cells appear to be a viable substitute for tissue biopsy. Blood cells have the ability to mirror the body's tissues and organs in health and disease; thus, we hypothesize that blood cells can indicate at the molecular level the presence of disease. Here we review microarray gene expression profiling of blood RNA for a number of different diseases. Sieving through gene expression molecular signatures has identified groups of genes characteristic of each and has identified biomarkers associated with specific diseases.

scholarly journals Pan-cancer machine learning predictors of primary site of origin and molecular subtype

2018 ◽  
Author(s):  
William F. Flynn ◽  
Sandeep Namburi ◽  
Carolyn A. Paisie ◽  
Honey V. Reddi ◽  
Sheng Li ◽  
...  
Keyword(s):  
Machine Learning ◽  
Molecular Subtypes ◽  
Molecular Subtype ◽  
Expression Profiles ◽  
Metastatic Cancer ◽  
Gene Expression Profiles ◽  
Primary Site ◽  
Cancer Types ◽  
Tissue Of Origin ◽  
Pan Cancer

ABSTRACTBackgroundIt is estimated by the American Cancer Society that approximately 5% of all metastatic tumors have no defined primary site (tissue) of origin and are classified as cancers of unknown primary (CUPs). The current standard of care for CUP patients depends on immunohistochemistry (IHC) based approaches to identify the primary site. The addition of post-mortem evaluation to IHC based tests helps to reveal the identity of the primary site for only 25% of the CUPs, emphasizing the acute need for better methods of determination of the site of origin. CUP patients are therefore given generic chemotherapeutic agents resulting in poor prognosis. When the tissue of origin is known, patients can be given site specific therapy with significant improvement in clinical outcome. Similarly, identifying the primary site of origin of metastatic cancer is of great importance for designing treatment.Identification of the primary site of origin is an import first step but may not be sufficient information for optimal treatment of the patient. Recent studies, primarily from The Cancer Genome Atlas (TCGA) project, and others, have revealed molecular subtypes in several cancer types with distinct clinical outcome. The molecular subtype captures the fundamental mechanisms driving the cancer and provides information that is essential for the optimal treatment of a cancer. Thus, along with primary site of origin, molecular subtype of a tumor is emerging as a criterion for personalized medicine and patient entry into clinical trials.However, there is no comprehensive toolset available for precise identification of tissue of origin or molecular subtype for precision medicine and translational research.Methods and FindingsWe posited that metastatic tumors will harbor the gene expression profiles of the primary site of origin of the cancer. Therefore, we decided to learn the molecular characteristics of the primary tumors using the large number of cancer genome profiles available from the TCGA project. Our predictors were trained for 33 cancer types and for the 11 cancers where there are established molecular subtypes. We estimated the accuracy of several machine learning models using cross-validation methods. The extensive testing using independent test sets revealed that the predictors had a median sensitivity and specificity of 97.2% and 99.9% respectively without losing classification of any tumor. Subtype classifiers achieved median sensitivity of 87.7% and specificity of 94.5% via cross validation and presented median sensitivity of 79.6% and specificity of 94.6% in two external datasets of 1,999 total samples. Importantly, these external data shows that our classifiers can robustly predict the primary site of origin from external microarray data, metastatic cancer data, and patient-derived xenograft (PDX) data.ConclusionWe have demonstrated the utility of gene expression profiles to solve the important clinical challenge of identifying the primary site of origin and the molecular subtype of cancers based on machine learning algorithms. We show, for the first time to our knowledge, that our pan-cancer classifiers can predict multiple cancers’ primary site of origin from metastatic samples. The predictors will be made available as open source software, freely available for academic non-commercial use.

Sign in / Sign up

Export Citation Format

Share Document