Systems Biology of Disease
June 19 - 23, 2017
Offered by the Institute for Systems Biology and the Center for Systems Biology
General Daily Structure
In the sections below you will find an overview for each day.
Concepts of systems thinking, networks, and systems properties will be described with examples. Various emerging technologies in systems biology will be explained. During the afternoon of the first day we will focus on using systems approaches for patient stratification. Clinical phenotypes of human diseases while appearing to be homogeneous pathologically, can in actuality be very heterogeneous in terms of the underlying molecular and genomic alterations. Pathologists have known this for quite some time, since patients with seemingly homogeneous pathology could result in very different patient outcomes. We will explain clustering analysis of heterogeneous data from large-scale biological datasets. As a group we will implement and experiment with various aspects of clustering analysis.
Quantifying Heterogeneity in Disease
An important source of heterogeneity among and within patients is at level of molecular and genomic characteristics of single cells or groups of cells, such as in the clonal lineages for a tumor. Recent technological advances offer the possibility to characterize molecular profiles from the patient’s tissue at single cell resolution. The quantification at the single cell resolution provides the data to more reliably diagnose patients and will ultimately lead to better treatments. Therefore, quantifying molecular signatures at single cell resolution is a vital step to define biomarkers and drug targets successfully. We will use publicly available single cell transcriptome data in the context of cancer genomics to learn how to use molecular profiles to characterize the heterogeneity both within and between patients.
In the context of this course a biomarker is considered a quantifiable molecular phenotype that can be measured and evaluated as an indicator of a biological state, pathogenic process or as a means to assess the therapeutic efficacy of a drug. Typically the features that make up the molecular signatures used to stratify patients into actionable subgroups or explain clinical phenotypes of interest are viable candidates for biomarkers. The development of biomarkers is a multi-step process involving discovering viable biomarkers, developing methods to screen them in non-invasive ways (blood or urine), and assessing their clinical implications. During the afternoon session, we will discuss and implement supervised machine learning approaches (classification) on a variety of large and current biological datasets.
Discovering Drug Targets
Improving the standard of care for persons with a disease is the end goal of systems biomedicine. The discovery of drug targets can be accomplished in a variety of ways but one underlying theme is the integration of prior information about the disease etiology. The integration of many different sources of information often leads to the construction of networks that can be mined for actionable hypotheses. As a case study, we will use a recent study that used a network based approach to discover combinations of drugs that have the potential to be used to treat tuberculosis. In this example we will demonstrate the power of network based approaches to layer information in such a way that it can be used to infer actionable predictions, e.g. combinations of drugs. Using unbiased integrative approaches with systems scale data it is possible to discover novel testable hypotheses.
Example Applications of Systems Biology
Friday morning will feature a mini-symposium to show trainees the many ways in which systems biology can be applied to biomedical studies. A lunch will be provided where trainees can discuss their insights and questions with the ISB faculty. On this day there will be plenty of time for discussion. Trainees are very strongly encouraged to think and discuss how applying systems biology approaches may enhance their own research. The symposium ends with a social event to which all of ISB is invited.
Before attending the course we strongly recommend that trainees take the ‘R Programming’ course from coursera.org
. You will need to make an account with courser (which is free), and then take the courser on ‘R Programming’ which takes a 4 weeks and is suggested to take 7-9 hours of time per week. Completion of this course is not required, but is highly recommended for those who are not familiar with R. Interested trainees may also take the edX course ‘Statistics and R for the Life Sciences’ to supplement their R skills.
To participate in this course each trainee will want to have a computer of their own. We will provide a list of software that the trainee is to have installed on their computer before coming to the course. A sample snippet of code will be used to determine if they have successfully installed the applications. If they don’t have them installed properly they can communicate the issues to us and we will help the as necessary. The goal being that they will come with computers ready to be used for the data analysis and exploration sections we have devised. We will also provide our old course laptops for those who do not have a laptop of their own or for those who have issues we cannot fix. These will be pre-installed with working versions of all software.