The estimated reading time for this post is 4 Minutes
For women in the U.S., breast cancer death rates are second only to lung cancer. It is estimated that approximately 30% of the cancers diagnosed in women will be breast cancer.
Breast cancer is a highly heterogeneous disease with different molecular subtypes making it even more complex. About 5-10% of breast cancers can be linked to gene mutations, which cause proteins to malfunction and disrupt the normal development of genes that increase the risk of cancer and are inherited from one’s parents. Surprisingly, about 85% of breast cancers occur in women who have no family history of breast cancer. These happen due to genetic mutations that occur as a result of the aging process and life in general, rather than inherited mutations.
Precision medicine is an emerging approach for disease treatment and prevention that takes into account individual variability in genes, environment, and lifestyle to more accurately predict which treatment and prevention strategies for a particular disease will work in which groups of people. Considering the complexity of breast cancer treatment, it is imperative researchers incorporate precision medicine in breast cancer diagnosis and treatment. In order to make this happen, a clearer understanding of the genomes of breast tumors is necessary.
A few decades ago, breast cancer classification systems were based on tumor response to endocrine therapy. In the last decade, new technologies for analyzing the genomic profiles of human tumors have substantially improved our understanding of the molecular classification of breast cancer. Currently, molecular classification of breast tumors is used alongside classical prognostic factors to predict tumor evolution and behavior, and to select specific treatments. This development improves diagnostic accuracy and enhances the ability to individualize therapy for breast cancer, thereby leading to direct implications for patient management. Unfortunately, computational studies, such as those that identify connections between genetic patterns and cancer characterization, have often proven to be difficult to reproduce. However, efforts are ongoing to characterize and improve the reproducibility of computational research.
Our first objective was to replicate the work conducted by Zhiyuan et al. Using the same data and procedures, we collected the intrinsic gene list upon which their single sample predictor was built. The intrinsic gene list is predicated on the idea that the gene expression pattern used to classify tumors succinctly summarizes characteristics of tumor samples taken from the same tumor, reflecting the fundamental differences of the tumors at the molecular level among cancer subtypes. We obtained similar results from conducting an independent study where our procedures matched the original experiment as closely as possible.
Currently, mankind has the capacity to capture and analyze biological information at the genetic level. Using this genomic data for disease prognosis and diagnosis is arguably one of the most important applications of this knowledge. However, several problems currently exist that impede the ability of researchers to effectively analyze these datasets. One such issue is the difficulty in obtaining sufficient sample data. Due to the costs – both financial and temporal – involved in the collection and processing of biological samples, most datasets are comprised of very few samples.
These few samples, however, typically contain many thousands of genetic features. This has led some researchers to explore methods for merging data sets from various studies into a single cohesive set. Given the differing objectives of different researchers, variations in sampling protocols, and no universal standard for data curation, unifying these data sets is not trivial. As with the original work, to generate a validation set we have merged four publicly available breast cancer expression datasets using Distance Weighted Discrimination (DWD) [8]. DWD is an analytical tool capable of merging disparate, multivariate data sets by making global adjustments designed to minimize biases inherent in those sets.
Finally, we have performed both hierarchical clustering and k-means clustering to seek interesting subpopulations of genes. Much of the previous analyses of gene expression data to classify breast tumors used hierarchical clustering. However, we will extend the work by trying different forms of hierarchical clustering and also k-means clustering to discover a distinctive molecular portrait of each tumor. We will show that we have been able to successfully achieve high reproducibility in identifying most of the subgroups previously identified in other studies.
Using the Kaplan-Meier survival analysis and Cox proportional-hazards model, we will compare the differences in outcomes and associations with other clinical parameters between each of the groups. While the hierarchical clustering technique will give us some distinct subgroups of genes, we show that k-means clustering performs better in identifying the distinct subgroups that are consistently predictive of a patient’s clinical outcomes, as evidenced by the prediction of Relapse-free survival and Overall survival of each identified group.
Did you like this example?
6845
285
#nursingstudent #nursingschool #nursing #nurse #nurselife #nursesofinstagram #nurses #rn #registerednurse #medical #studentnurse #doctor #nursepractitioner #nursesrock #medicalstudent #medicine #nclex #lpn #futurenurse #nursingschoolproblems #nurseproblems #healthcare #nursinglife #scrublife #nursehumor #hospital #medicalschool #medstudent #medschool #nursingstudents
We can help you complete this assignment or another one similar to this. Just hit "Order Now" to get started!
