Tumor Classification from Gene Expression Data: A Coding-Based Multiclass Learning Approach

Breast cancer outcome can be predicted using models derived from gene expression data or clinical data. Only a few studies have created a single prediction model using both gene expression and clinical data. These studies often remain inconclusive regarding an obtained improvement in prediction performance. We rigorously compare three different integration strategies (early, intermediate, and late integration) as well as classifiers employing no integration (only one data type) using five classifiers of varying complexity. We perform our analysis on a set of 295 breast cancer samples, for which gene expression data and an extensive set of clinical parameters are available as well as four breast cancer datasets containing 521 samples that we used as independent validation.mOn the 295 samples, a nearest mean classifier employing a logical OR operation (late integration) on clinical and expression classifiers significantly outperforms all other classifiers. Moreover, regardless of the integration strategy, the nearest mean classifier achieves the best performance. All five classifiers achieve their best performance when integrating clinical and expression data. Repeating the experiments using the 521 samples from the four independent validation datasets also indicated a significant performance improvement when integrating clinical and gene expression data. Whether integration also improves performances on other datasets (e.g. other tumor types) has not been investigated, but seems worthwhile pursuing. Our work suggests that future models for predicting breast cancer outcome should exploit both data types by employing a late OR or intermediate integration strategy based on nearest mean classifiers.

In order to understand even the simplest cellular processes, we need to integrate proteomic, gene expression and other biomolecular data. To date, most computational approaches aimed at integrating proteomics and gene expression data use direct gene/protein correlation measures. However, due to post-transcriptional and translational regulations, the correspondence between the expression of a gene and its protein is complicated. We apply a multivariate statistical method, co-inertia analysis (CIA), to visualise gene and proteomic expression data stemming from the same biological samples. Principal components analysis or correspondence analysis can be used for data exploration on single datasets. CIA is then used to explore the relationships between two or more datasets. We further explore the data by projecting gene ontology (GO) information onto these plots to describe the cellular processes in action. We apply these techniques to gene expression and protein abundance data from studies of the human malarial parasite life cycle and the NCI-60 cancer cell lines. In each case, we visualise gene expression, protein abundance and GO classes in the same low dimensional projections and identify GO classes that are likely to be of biological importance.

The Cancer Genome Atlas (TCGA) has made available comprehensive molecular profiles including gene expression using RNA-seq for many human tumor types and, thereby, provided a great opportunity to use pan-cancer gene expression patterns to identify unique features that can classify tumor types. Those features may serve as biomarkers for tumor diagnosis and potential targets for drug development. Because gender differences in cancer susceptibility are one of the most consistent findings in cancer epidemiology, knowing whether the distinguishing features differ between males and females for the same tumor types might enhance their utility as biomarkers. Our goals are: 1) to identify a set of genes whose expression levels can classify pan-cancer tumor types when gender is ignored; and 2) to identify analogous sets of genes for pan-cancer classification in sex non-specific tumors from men and from women separately. Here we analyzed the RNA-seq expression data for 9,096 TCGA tumor samples from 31 tumor types; we selected features for sample classification using a genetic algorithm, GA/KNN, which we developed. We randomly assigned half of all samples into a training set and half into a test set, proportionally allocating samples from each tumor type. We used the training set to obtain 2,000 near-optimal classifiers (each set consisting of 20 genes) from repeated GA/KNN runs. We subsequently applied each resultant near-optimal classifier to the test set. The predicted class was then compared with the true class to calculate both training- and testing-set prediction accuracies. To see if sex non-specific tumors (those that are not related to reproductive organs) differ between males and females, we repeated the analysis for 24 sex non-specific tumor types separately for males and for females. In the analysis ignoring gender, we could accurately classify 88.7% training samples and 88.1% of the testing samples on average. The top 20 genes that separated the 31 tumor types were TCF21, TBX5, EMX20S, EMX2, PA2G4P4, HNF1B, NACA2, SFTPA1, ATP5EP2, PTTG3P, FTHL3, ANXA2P3, GATA3, NAPSA, SFTA3, HSPB1P1, HOXA9, IGBP1P1, RPL19P12, and SFTPB. Not surprisingly, we achieved similar average classification accuracies for testing samples (88.2% for males and 88.4% for females). Seventy four of the 100 top-ranked genes were common between males and females. The main reasons for less than total overlap between males and females are: stochastic nature of the algorithm, training/testing partitioning (only done once) and some intrinsic differences between genders. For example, BNC1, FAT2, and MSX2 ranked 36, 84, and 94 in males but 461, 603, and 1085 in females. Conversely, HPN, POU3F3, and FOXA1 ranked 75, 78, and 91 but 421, 355, and 349, respectively. It remains unclear whether genes likes those play any role in differential cancer susceptibility between genders. In conclusion, we were able to select many sets of 20 genes that can correctly classify ~88% of the tumor samples from 31 different types for a validation set using RNA-seq gene expression alone. This accuracy is remarkable given the number of the tumor types involved. This result was largely replicated in tumors between genders. We were also identified a few genes that might play a role in gender differences among sex non-specific tumors. Citation Format: YuanYuan Li, David M. Umbach, Leping Li. A comprehensive genomic pan-cancer analysis comparing males and females using The Cancer Genome Atlas gene expression data. [abstract]. In: Proceedings of the AACR Precision Medicine Series: Targeting the Vulnerabilities of Cancer; May 16-19, 2016; Miami, FL. Philadelphia (PA): AACR; Clin Cancer Res 2017;23(1_Suppl):Abstract nr A46.

The advent of gene chips has led to a promising technology for cell, tumor, and cancer classification. We exploit and expand the methodology of recursive partitioning trees for tumor and cell classification from microarray gene expression data. To improve classification and prediction accuracy, we introduce a deterministic procedure to form forests of classification trees and compare their performance with extant alternatives. When two published and commonly used data sets are used, we find that the deterministic forests perform similarly to the random forests in terms of the error rate obtained from the leave-one-out procedure, and all of the forests are far better than the single trees. In addition, we provide graphical presentations to facilitate interpretation of complex forests and compare our findings with the current biological literature. In addition to numerical improvement, the main advantage of deterministic forests is reproducibility and scientific interpretability of all steps in tree construction.

In the healthcare research community, Internet of Medical Things (IoMT) is transforming the healthcare system into the world of the future internet. In IoMT enabled Computer aided diagnosis (CAD) system, the Health-related information is stored via the internet, and supportive data is provided to the patients. The development of various smart devices is interconnected via the internet, which helps the patient to communicate with a medical expert using IoMT based remote healthcare system for various life threatening diseases, e.g., brain tumors. Often, the tumors are predecessors to cancers, and the survival rates are very low. So, early detection and classification of tumors can save a lot of lives. IoMT enabled CAD system plays a vital role in solving these problems. Deep learning, a new domain in Machine Learning, has attracted a lot of attention in the last few years. The concept of Convolutional Neural Networks (CNNs) has been widely used in this field. In this paper, we have classified brain tumors into three classes, namely glioma, meningioma and pituitary, using transfer learning model. The features of the brain MRI images are extracted using a pre-trained CNN, i.e. GoogLeNet. The features are then classified using classifiers such as softmax, Support Vector Machine (SVM), and K-Nearest Neighbor (K-NN). The proposed model is trained and tested on CE-MRI Figshare and Harvard medical repository datasets. The experimental results are superior to the other existing models. Performance measures such as accuracy, specificity, and F1 score are examined to evaluate the performances of the proposed model.

A reliable and precise classification of tumors is essential for successful diagnosis and treatment of cancer. cDNA microarrays and high-density oligonucleotide chips are novel biotechnologies increasingly used in cancer research. By allowing the monitoring of expression levels in cells for thousands of genes simultaneously, microarray experiments may lead to a more complete understanding of the molecular variations among tumors and hence to a finer and more informative classification. The ability to successfully distinguish between tumor classes (already known or yet to be discovered) using gene expression data is an important aspect of this novel approach to cancer classification. This article compares the performance of different discrimination methods for the classification of tumors based on gene expression data. The methods include nearest-neighbor classifiers, linear discriminant analysis, and classification trees. Recent machine learning approaches, such as bagging and boosting, are also considered. The discrimination methods are applied to datasets from three recently published cancer gene expression studies.

Gene expression based cancer classification

Diagnostic procedures concerning soft tissue tumors are an important issue in oncology, underlined by the need for correct, timely diagnosis that will buy time toward some leads on an effective treatment plan. Traditional diagnostic techniques, though reliable most of the time, are often invasive and time-consuming. Machine learning techniques are explored to establish the possibility of using them in the automated diagnosis of soft tissue tumors, therefore improving accuracy and reducing diagnosis time. We will make use of histopathological images and clinical data in a dataset, running a number of machine learning algorithms for the identification and classification of tumor types, including convolutional neural networks. Propose an improved approach to significantly improve diagnostic accuracy compared to the conventional methods, thus having the potential to assist pathologists in more informed decision-making. The potential integration of machine learning within diagnostic workflow is prone to revolutionize oncological principles by the offer of non-invasive, fast, and highly accurate alternatives for the early detection and classification of soft tissue tumors. Our results underline the feasibility and advantages that could be brought into reality by an automated diagnosis system and open ways toward developing more advanced, available tools in cancer diagnosis.

Detecting brain tumors early on is critical for effective treatment and life-saving efforts. The analysis of the brain with MRI scans is fundamental to the diagnosis because it contains detailed structural views of the brain, which is vital in identifying any of its abnormalities. The other option of performing an invasive biopsy is very painful and uncomfortable, which is not the case with MRI as it is free from surgically invasive margins and pieces of equipment. This helps patients to feel more at ease and hasten the diagnostic procedure, allowing physicians to formulate and practice action plans quicker. It is very difficult to locate a human brain tumor by manual because MRI scans produce large numbers of three-dimensional images. Complete applicability of pre-written computerized diagnostics, affords high possibilities in providing areas of interest earlier through the application of machine learning techniques and algorithms. The proposed work in the present study was to develop a deep learning model which will classify brain tumor grade images (BTGC), and hence enhance accuracy in diagnosing patients with different grades of brain tumors using MRI. A MobileNetV2 model, was used to extract the features from the images. This model increases the efficiency and generalizability of the model further. In this study, six standard Kaggle brain tumor MRI datasets were used to train and validate the developed and tested model of a brain tumor detection and classification algorithm into several types. This work consists of two key components: (i) brain tumor detection and (ii) classification of the tumor. The tumor classifications are conducted in both three classes (Meningioma, Pituitary, and glioma) and two classes (malignant, benign). The model has been reported to detect brain tumors with 99.85% accuracy, to distinguish benign and malignant tumors with 99.87% accuracy, and to type meningioma, pituitary, and glioma tumors with 99.38% accuracy. The results of this study indicate that the described technique is useful in the detection and classification of brain tumors.

BackgroundA complete understanding of the regulatory mechanisms of gene expression is the next important issue of genomics. Many bioinformaticians have developed methods and algorithms for predicting transcriptional regulatory mechanisms from sequence, gene expression, and binding data. However, most of these studies involved the use of yeast which has much simpler regulatory networks than human and has many genome wide binding data and gene expression data under diverse conditions. Studies of genome wide transcriptional networks of human genomes currently lag behind those of yeast.ResultsWe report herein a new method that combines gene expression data analysis with promoter analysis to infer transcriptional regulatory elements of human genes. The Z scores from the application of gene set analysis with gene sets of transcription factor binding sites (TFBSs) were successfully used to represent the activity of TFBSs in a given microarray data set. A significant correlation between the Z scores of gene sets of TFBSs and individual genes across multiple conditions permitted successful identification of many known human transcriptional regulatory elements of genes as well as the prediction of numerous putative TFBSs of many genes which will constitute a good starting point for further experiments. Using Z scores of gene sets of TFBSs produced better predictions than the use of mRNA levels of a transcription factor itself, suggesting that the Z scores of gene sets of TFBSs better represent diverse mechanisms for changing the activity of transcription factors in the cell. In addition, cis-regulatory modules, combinations of co-acting TFBSs, were readily identified by our analysis.ConclusionBy a strategic combination of gene set level analysis of gene expression data sets and promoter analysis, we were able to identify and predict many transcriptional regulatory elements of human genes. We conclude that this approach will aid in decoding some of the important transcriptional regulatory elements of human genes.

Diagnostic performance of radiomics and machine learning algorithms in differentiating grade 2-3 gliomas from glioblastomas among adult-type diffuse gliomas.

Improve clinical feature-based bladder cancer survival prediction models through integration with gene expression profiles and machine learning techniques

Pictorial essay on selected new, emerging, and reclassified entities in the 2020 WHO classification of soft tissue tumors.

Disease Prediction From Various Symptoms Using Machine Learning

Due to the complexity of the images and dearth of anatomical models, it is highly difficult to accurately represent the various deformations in each component of the medical images. In recent years, a significant number of children and adults have affected from brain tumors, which is one of the most terrible types of disease affects the people around the world. Moreover, the Magnetic Resonance Imaging (MRI) based brain tumor detection is one of a significant study area in the field of medical imaging. Since, the use of computerized methods aids in the detection and treatment of disease by the medical professionals. The development of an automated method for the accurate detection and classification of tumors from brain MRIs. In this framework, a tanh normalization process is used to smooth out the input brain MRIs with less noise artefacts and improved quality. Then, a group feature extraction model is used to extract the relevant features from the normalized image, which includes both Speeded Up Robust Features (SURF) and Grey Level Co-occurrence Matrix (GLCM) features. The Water Chaotic Fruitfly Optimization (WChFO) method is used to identify the best features for increasing the speed of classifier training and testing processes with less time. Moreover, a Deep Recurrent Neural Network (DRNN) model is used to classify the type of brain tumor for accurate early diagnosis and treatment. The most well-known benchmarking datasets, like BRATS and Kaggle, employed for analysis in order to assess the effectiveness and results of the proposed brain tumor diagnosis system. By using the proposed WChFO-DRNN technique, the accuracy of the tumor detection system is increased to 99.2% with the sensitivity, specificity of 99% and time consumption of 0.2s.