improved responsiveness, lowered
costs, and increased reliability.
Although many organizations are
attempting to implement efficient
streamlined systems, ever-increasing
regulatory pressures are forcing the
pharmaceutical industry to lag behind
other industries in adopting new technologies, despite their urgent need to
Faced with cost-savings and economic
pressures, organizations are starting to
look again at how they can drive R&D
efficiency and, in particular, many R&D
organizations are considering the use
of automated technologies to free up
capacity and make the flow of information from hypothesis to product faster.
But, as a data-intensive industry, they
are facing several challenges.
Pharma's Big Data Problem
“Big data” is a hot buzzword, and understandably so. Along with analytics,
artificial intelligence (AI), and machine
learning (ML), big data is part of a
broader wave of technology sweeping
through the industry that could lead
to the ultimate form of automation.
But, if pharma wants to ride the wave
and reap the benefits, there are a few
roadblocks that need to be overcome.
Consider the “ 3 Vs” of big data:
volume, velocity, and variety. It is
clear that there is a huge volume
of data being produced across the
pharmaceutical R&D value chain, and
with a great velocity of change, from the
reams of data being produced through
clinical trials to experimental data
being generated by more and more
sophisticated equipment and sensors.
However, it’s perhaps the variety of data
produced that provides the biggest
problem. Being able to integrate this
data is crucial for organizations trying to
gain a wider insight and maximize the
value of the data being generated.
A Legacy Challenge
Having reliable, well-linked data is
essential for any organization looking to
automate and benefit from big data, AI,
and ML technology trends. However,
pharmaceutical companies have to deal
with a vast array of legacy systems, each
containing disparate data and a lack of
standardized data formats.
The issue they face is that these
systems have been designed and
implemented to answer specific
questions of the data being
collected, but therefore limiting
potential other insights that can be
found from interrogating this data.
In order to take advantage of more
sophisticated analysis methodologies,
and lead to the potential for wider insights, these legacy data silos need to
be broken down and the data linked
together in a clean way.
Many legacy systems were built on
“old” technology, making it very
difficult and costly to extract the data
from these systems into an integrated
environment where advanced analytics can be performed. This has led
to a skills gap in organizations, which
now require data scientists capable of
connecting this “noisy” data, extracting the value and presenting it to
stakeholders to allow them to make
While there is still a great deal to be
done to create standard data formats
within the pharmaceutical industry,
vast inroads already have been made.
Examples include the work being
performed by the Pistoia Alliance to
standardize scientific data capture
with initiatives such as their Unified
Data Model (UDM) project, creating
a data format for the storage and
exchange of experimental information,
or their HELM project to create a
single notation that can encode the
structure of all biomolecules. Other
data standards such as the emerging
Allotrope ADF and well developed
AnIML standards are helping ease
integration of analytical instrument data.
These, among others, are making
the journey to an integrated and
automated drug discovery process a
closer reality. Although there are still
significant hurdles for pharmaceutical organizations, we are beginning
to see a change of mindset in the
industry with organizations such as
GlaxoSmithKline putting a stronger
focus on automation, analytics, and AI
to drive drug discovery.
What Does the Future Hold?
When looking to the future, the only
certainty is that the volumes of data
being produced through the Pharma
R&D value chain is going to rise—
and the technology that allows data
to be interrogated and lead to new
breakthroughs also will continue to
develop at pace.
The only remaining question is this:
Can an industry with such regulatory sensitivity and data silos move fast enough
to keep up with the technology?
About the Author
Rory Quinn is Solution
Consultant at IDBS. The
company was acquired by
the Danaher Group and
joined their Life Sciences
platform in late October 2017.
While there is still a great deal to be done
to create standard data formats within
the pharmaceutical industry, vast inroads
already have been made.