Approaches to Web Development for Bioinformatics

Introduction

One of the driving forces behind research in biology and medicine over the past decade or two has been the availability of huge amounts of DNA and protein data.  The Human Genome Project¹ was one of the major initiatives.  It led to massive improvements in the efficiency of DNA sequencing, and inspired numerous other genome projects.  However, you do not have to be the US National Institute of Health (NIH) to create a bioinformatics portal.  Smaller organizations may have many reasons for and building a bioinformatics portal to collaborate with others, including sharing data and research results.

Given that you have an idea for analyzing or presenting data in a particular was, a complete bioinformatics web application depends of these basic pieces, which is what this article is all about:

1. A source of data. This may be your own organization's data or data from a public source.
2. An application programming language to access and analyze the data. With this you need to know how to access and process the data in the first step.
3. A web application platform to provide a HTML user interface for your data and analysis results. This enables you to present the data and model from the previous step to your users and allow them to interact with it.
4. Optionally, a data store, such as a relational database, to store results or user's data. If your users are to save any data for use in a future visit you will need this.
5. Optionally, you would reuse software tools and libraries developed by others. This can save you time and may enable you to do things that you don't have the expertise to do yourself.

There are a number of bioinformatics databases that are publicly available and, in addition, make the databases and software tools themselves available for building your own web user interfaces.  The databases include:

• Created by the National Center for Biotechnology Information (NCBI) at the US National Institute of Health (NIH), the GenBank ² database is a DNA sequence database with genetic data from humans from the Human Genome Project and also other organisms. 
• The European Bioinformatics Institute³ has similar genetic databases and tools for working with the data. 
• The Swiss Institute of Bioinformatics (SIB) hosts the ExPASy (Expert Protein Analysis System) proteomics server ⁴, which includes a protein knowledge base and tools.
• The Kyoto Genes and Genome (KEGG) database⁵ is another genetic information database but also has pathway, ligand, and drug information and has a web services API to access the database.
• The Institute for Genomic Research (TIGR)⁶ also hosts a genome sequence database also hosts several genome databases.
• Ensembl¹⁹ - a joint project between EMBL - EBI and the Sanger Institute to maintain genome maps for a number of organisms
• The Broad Institute²¹ is a joint institute between MIT, Harvard and their affiliated hospitals and the Whitehead Institute.  It hosts genome, genetic variation, inhibitory RNA, and chemical biology databases, and cancer datasets.

When developing bioinformatics software, it can be a challenge to be an expert in biology and in software engineering at the same time.  This is made worse by the many different programming platforms and technologies used in bioinformatics.  For this reason, I have included some basic material to get something basic running on each platform to get an appreciation of what the issues will be and then provide pointers to further resources.  My goal is to use this as an aid to get started and consider the choices from so many different software technologies. The following table summarizes the programming languages, web application platforms, and bioinformatics tools and libraries discussed in this article as well as their relative advantages and disadvantages.

Many of the tools described are open source.  This can present a challenge to some commercial organizations where ownership of intellectual property is very important.  See the article Open Source Software and Documentation for Chemistry and Biomedical Sciences on this web site for more details on open source.

After looking at each we will find that, despite the existence strengths and weaknesses, all the web application platforms discussed have the fundamental pieces needed for bioinformatics applications and, at a fundamental level, have comparatively similar capabilities. Given that, there will be many bioinformatics applications written on different platforms. The next question is how do you integrate them together to provide users with something they can handle? The section Application Integration discusses these questions.

See the page Bioinformatics Resources on this web site for a list of popular and user suggested online resources.

Please send ideas and opinions by email at alexamies@gmail.com.