Zack Lynch is author of The Neuro Revolution: How Brain Science Is Changing Our World (St. Martin's Press, July 2009).
While bioinformatics isn't likely to create any new software stars, neuroinformatics will. The reason is simple: complexity. As I mentioned recently, the data about a person's genome can already fit on an ipod, yet the data about one's brain will require petabytes, if not exabytes, of storage capacity.
How much is a petabyte? One example is the Internet Archive Wayback Machine that contains approximately 1 petabyte of data and it has been archiving almost every webpage created since 1993.
Along with government initiatives like the human brain project there are also several small companies targeting neuroinformatics like Australia's Brain Resource Company (BRC), San Diego's Neurome, and Chicago's MIICRO.
Here is an overview of what BRC is up to (courtesy of Psychscape)
BRC has has set up the world's first standardized international database on the human brain. BRC already has a database of over 1,000 normative subjects and over 500 clinical subjects and still growing. This collaboration of scientists and technology partners (such as IBM) gathers information into a neuroscience database which includes demographic, neuropsychological (cognitive), electrical brain-body function, sMRI, fMRI, genetic and lifestyle data along with function, structure and genetics of patients' brains. A patient who is referred to the database (by over 50 researchers from around the globe), first enters data online - this consists of demographic data such as age, gender, eating and drinking habits, early childhood experiences. The patient tben goes into one of the BRC labs to undergo various tests, such as MRIs and EEGs
According to the BRC website, researchers then use a tool called the "Matrix" that allows comparisons between various elements in the database. "It consists of 245 x 245 correlations, with 8 layers of age, (each with 3 parameters) totalling over 1.4 Million cells of data. This enormous amount of information is powerfully summarised by automated colouring of cells based on significance levels. At a glance widespread patterns in the data can be seen as patches of colour. To further investigate such hotspots the matrix can be crossed referenced on all three dimensions (correlates of the column variable, correlates of the row variable, and through age groups/covariation) to explore possible confounds, interaction and causality."
One of the goals of the BRC is to allow rapid comparisons of a patient profile against the normative data with the goal of predicting a response to particular drugs or anticipate a side effect to a specific intervention. Science has been chasing the ability to predict a personal response to any clinical intervention. Who will respond and who will not respond is extremely valuable information to the pharmaceutical industry as well as to clinicians.