A New Era in Brain Research: Applying Big Data approaches and Machine Learning algorithms
Huge advances in data-driven computational tools can help unlock the brain’s complexities, empowering clinicians to improve decision making and personalize treatment of mental illness.
Data is everywhere – it’s all around us. From traditional paper or digital sources – we are generating and collecting data in enormous amounts. Computational tools, such as Machine Learning algorithms, now enable us to automatically analyze all this structured and unstructured “Big Data” to discover new hidden secrets with huge value.
Its use in business is already second nature. From Facebook’s ability to tailor our news-feeds, to Google’s ability to fit the best advertisement to every user, Big Data and prediction algorithms are constantly personalizing our online experiences.
The power of these tools brings huge opportunities to the medical world. Scientists and doctors now have access to a constant source of clinical, genetic and environmental Big Data for ongoing discovery and analysis that simply wasn’t previously available!
Furthermore, the analysis of all this Big Data can help unlock the brain’s mysteries to reveal new insights with great clinical value. This new information, can empower scientists and doctors to improve medical recommendations and personalize treatment of mental illness.
The need for new approaches in brain research
The brain is the most complex organ in the body. We remain unsure of its vastness and complexity despite detailed research into understanding the anatomical intricacies of brain functioning. We are also still ignorant of how the brain coordinates all of its activities and the delicate interaction of different factors in driving our behaviors, thoughts and sense of self.
As a scientist, I believe that brain research must take advantage of these new "intelligent tools" to enable transformative innovation in the treatment of psychiatric disorders. Only by shifting our approach through the use of deep analysis by means of advanced methodologies, can we better understand the brain’s complexity and empower clinicians to make more informed decisions.
One important aspect, where these data-driven computational tools can help, is the convoluted and important role of the environment and genetics on psychiatric disorders. This profound vision was my drive to establish the Taliaz personalized medicine analytics company.
The interaction between genetic and environmental factors and psychiatric disorders
The impact of the environment on gene expression
To date, research into psychiatric disorders has focussed mainly on associations with single genetic factors, with the great strides made by the Genome-Wide Association Studies (GWAS) in the last decade. However, these approaches have not incorporated and connected the interaction of genetic and environmental factors on brain development and behavior, and its associated psychiatric disorder and treatment.
A good example of why these factors matter can be seen in research by McGowan et al (2009)[i]; where the environmental experience of childhood abuse may lead to specific epigenetic changes, which could affect the hypothalamic-pituitary-adrenal (HPA) stress responses, and may increase the behavioural risk of suicide!
Clearly, understanding the wider impact of the environment on gene expression is crucial to understand the background disorder of each patient for effective treatment delivery.
The opposing effects of gene expression in different brain regions
Effective targeted therapy based on gene expression biomarkers must understand the multifaceted and opposing effects of gene expression in distinct brain regions.
This issue was clearly highlighted during my PhD researching brain-derived neurotrophic factor’s (BDNF) complex relationship with depression. In one of my studies[ii], we showed opposing roles for BDNF in the hippocampus and the ventral tegmental area, and reciprocal interactions with one and the other in these two distinct brain regions. BDNF’s opposing multifaceted role within different brain regions made targeting it as a biomarker for mood disorders highly complex!
What we needed was new approaches to help us delve deeper into the complexities of the brain as a means to improve the treatment of depression and associated psychiatric disorders. This is one of the research questions we are addressing in Taliaz.
Big Data and Machine Learning methods can transform the way we treat psychiatric disorders
The benefits of analyzing Big Data using Machine Learning
Big Data and Machine Learning algorithms are continuously being used to improve and personalize our life journey by automatically learning from experience.
For example, the giant U.S retailer, Target, gained the ability to predict motherhood through its data-mining program! By looking at the contents of their customers’ shopping baskets, they could spot customers who they thought were likely to be expecting and begin targeting promotions for nappies (diapers), cotton wool and so on. The prediction was so accurate that Target made the news by sending promotional coupons to families who did not yet realise (or who had not yet announced) they were pregnant! To learn more, read here.
Yet to date, applications in healthcare and specifically for mental health have remained relatively limited.
The challenges and benefits of applying Machine Learning algorithms in healthcare
Their application is no easy feat. When applying a Machine Learning method, data samples constitute the basic components. Every sample is described with multiple features and every feature consists of different types of values. The challenge of any prediction algorithm is to select the right combination of features that will predict the clinical outcome, e.g., history of childhood abuse and specific genetic components associated with stress response.
The good thing is that the accumulation of Big Data means we now have large numbers of parameters to find the right markers. We also have access to new genetic and clinical data, due to the advances in Next Generation Sequencing (NGS) technologies and Electronic Medical Records (EMR). This new information means we can now perform research and analysis that was simply not feasible before!
A new era in brain research
Advances in computer science, mathematical analysis, genetic sequencing, and brain imaging means we are entering into a new era of brain research: “The intelligent era”. Where through data-driven analytical methods, brain research is conducted in more sophisticated ways to help us better understand the brain’s complexities to deliver true personalized medicine.
With these new insights will come new approaches and solutions to help us more effectively treat psychiatric disorders. Furthermore, the analysis of the environmental and clinical impact for the first time, alongside genetic understanding, has the potential to transform brain research and mental healthcare!
We now have the technology to understand the brain’s interconnected intricacies, rather than just specific and anecdotal associations.
I am a great believer that these "intelligent tools" will help us redesign the way we as scientists can support clinicians to predict therapies for patients.
Join me in my next blog, where I will discuss how applying Big Data approaches on genetics can further help us unravel the complexity of mental illness.
[i] McGowan PO et al. (2009) Epigenetic regulation of the glucocorticoid receptor in human brain associates with childhood abuse. Nature Neuroscience 12:342–348.
[ii] Taliaz D, Nagaraj V, Haramati S, Chen A, and Zangen A. Altered Brain-Derived Neurotrophic Factor Expression in the Ventral Tegmental Area, but not in the Hippocampus, Is Essential for Antidepressant-Like Effects of Electroconvulsive Therapy. Biol Psychiatry; 74:305–312.