HPC and Grid Computing for Integrative Biomedical Research

Abstract

Integrative biomedical research projects query, analyze, and integrate many different data types and make use of datasets obtained from measurements or simulations of structure and function at multiple biological scales. With the increasing availability of high-throughput and high-resolution instruments, the integrative biomedical research imposes many challenging requirements on software middleware systems. In this paper, we look at some of these requirements using example research pattern templates. We then discuss how middleware systems, which incorporate Grid and high-performance computing, could be employed to address the requirements.

Re-Using Workflow Fragments across Multiple Data Domains

“…Thus, they also need to use approaches from many different disciplines to analyze data from multiple domains. Examples such as this also abound in Biology [5], particularly in Proteomics and Genomics.…”

“…This kind of re-use and re-purposing of workflow fragments across different data domains can be generalized to other scientific fields and allows scientists to link across different disciplines [2], [5]. For example, in geosciences, researchers observe the surface of the Earth at criticial points and examine moisture levels from above and below.…”

“…Scientific workflows help computational research in many different disciplines by consolidating heterogenous codebases and programs written in many different languages [1], [2], [3], [4], [5], [6]. Such workflows, designed by domain experts in their own fields, may also be of great utility to scientists in other disciplines; in fact, sites like http://www.myexperiment.…”

“…To facilitate this export of workflows and workflow fragments, we utilize the Wings workflow system, which was developed to assist scientists in managing complex computations [3] and has been used in several large-scale distributed scientific applications [5]. Wings uses semantic workflow representations that capture the requirements and semantic constraints of individual steps and datasets explicitly, as well as workflow reasoning algorithms to generate and validate possible combinations of workflow components systematically.…”

Large-scale multimedia content analysis using scientific workflows

“…Our approach is based on scientific workflows, which capture end-to-end methods that combine fragments developed by different people with expertise in different aspects of the task. Workflows consolidate heterogeneous codebases and programs written in many different languages [4,5,6]. Such workflows, designed by domain experts in their own fields, may also be of great utility to scientists in other disciplines.…”