Applied Data Mining: From Biomarker Discovery to Decision Support Systems
This chapter provides an overview of emerging bioinformatics methods for the biomarker discovery process and medical decision support. It introduces study design consideration and bioanalytic concepts for generating biomedical data, followed by various da
- PDF / 125,125 Bytes
- 12 Pages / 504.567 x 720 pts Page_size
- 3 Downloads / 227 Views
10
M. Osl, M. Netzer, S. Dreiseitl, and C. Baumgartner
Abstract
This chapter provides an overview of emerging bioinformatics methods for the biomarker discovery process and medical decision support. It introduces study design consideration and bioanalytic concepts for generating biomedical data, followed by various data mining and information retrieval procedures such as feature selection, classification as well as statistical and clinical validation. The reviewed methods are illustrated by real examples from preclinical and clinical studies, and the application in medical decision making is discussed. This chapter is anticipated to address to those with a bioinformatics background as well as biomedical researchers who are interested in the application of computational methods in biomarker discovery and medical decision making.
10.1
Introduction
With sequencing and profiling the complete human genome and a broad functional repertoire of the human proteome and metabolome, biomedical research is profoundly altering into a variety of not-yet-seen and unexpected directions by replacing established methods, tools, and standard procedures by novel, revolutionary applications in clinical medicine. The next-generation high-throughput sequencing machines are the keys to uncovering thousands of not yet identified, disease modulating genes that allow for
C. Baumgartner (*) Institute of Electrical and Biomedical Engineering, University for Health Sciences, Medical Informatics and Technology, Hall in Tirol, Austria e-mail: [email protected]
speeding up and automating the yet limited procedures with the existing technologies to prepare and sequence DNA samples. Consequently, faster sequencing permits faster access to the genetic cause of a disease, and with the complementary power of next-generation sequencing technologies, biochemistry and bioinformatics for processing the enormous data pool this procedure will get dramatically accelerated. Hence, the way from sample collection to identifying putative genetic signatures is expected to become shorter. This means that researchers can immediately begin to develop gene and drug therapies for that particular disease, and clinicians will benefit from novel therapies that are available more quickly for patient management and treatment (Meyerson et al. 2010; Ding et al. 2010). In the same manner, modern genomic, proteomic, and metabolic profiling technologies have
Z. Trajanoski (ed.), Computational Medicine, DOI 10.1007/978-3-7091-0947-2_10, # Springer-Verlag Wien 2012
173
174
M. Osl et al.
been established very recently and are now routinely available in clinical labs for assisting in comparative genomics analysis, or in detecting low-abundance and low-weight biomarkers associated with disease or disease-related pathways (Baumgartner et al. 2008). Diverse mass spectrometry (MS) instrumentations coupled with traditional separation techniques such as GC, LC, or HPLC are the key technologies—showing high sensitivity and structural specificity—to successfully aid
Data Loading...
