In the current economic climate there is more pressure than ever on the pharma industry to provide safe and effective medicines in as short a timescale as possible.

While the regulatory framework in terms of development milestones remains the same, drug developers are continuously looking to new techniques and technologies as adjuvants to help optimise the process. Previously, the use of microarray technology has been primarily confined to target identification and genetic testing; However, more recently, researchers have begun applying this technology at various stages of the drug development process in order to gain further insights as to the effects (both good and bad) of their compounds. Enrichment of the results and subsequent pathway analysis can further augment the information obtained and create a genomic signature for the disease of interest.

Microarray analysis is being used in Phase I healthy volunteer studies to simply determine if the drug is effecting the cellular pathways of interest. Conversely, and using the same samples, any off target effects of compounds could be assessed using the same technology. Previous studies have quite clearly shown that genomic signatures associated with toxicity can be detected long before the side effect phenotype is observed (and at lower doses).

By using blinded microarray analysis in Phase II studies one can differentiate between responders and non-responders and genomic signatures can actually be used to stratify patients ensuring the correct cohort are included in the trial in the first instance giving any new compound the best chance of success.

The data generated by microarray analysis brings a new set of problems as outputs are complex and yield highly diverse data sets which, at first look, can show no evidence of patterns or trends. The sheer volume of large clinical datasets can also be daunting. Phase II trials are especially challenging as generally this will be the first data set generated in a disease-suffering, outbred population. Such demographics create genomic "noise" further complicating the analysis.

Fios Genomics specialise in the analysis of highly complex genomic data sets. Our innovative computational solutions allow any patterns or trends to be revealed in even the most heterozygous datasets. Only by parallelising the data analysis in large data sets, using the most up-to-date tools and by applying robust statistical filtering can the results be unravelled and the biology of the experiment or trial be revealed. Our approaches allow analyses to be performed to a broader scope and deeper level than conventional methods allowing confident decision making and minimising the chances of developmental surprises later on.

The results obtained from a thorough bioinformatics analysis of clinical genomic data can be used in various ways such as development of companion diagnostics or indeed in the field of repositioning in the case of "off" target effects. As a result researchers can fully investigate the properties of their candidates, make informed decisions and get the most from their assets.