Machine learning, deep learning, and pre-competitive data analysis will deliver tangible benefits to patients

London, 29 March, 2017 – The Pistoia Alliance, a global, not for profit alliance that works to lower barriers to innovation in life sciences R&D, is calling upon the industry to improve collaborative efforts to use patient data to its full effect. In a series of keynote speeches delivered at The Pistoia Alliance’s annual member conference in London, speakers from Amgen, Accenture and AstraZeneca, discussed the need to more closely connect outcomes data with the R&D process – to help pharmaceutical companies focus their research efforts and deliver real benefits to patients. Building machine learning and deep learning systems, and incorporating data from therapeutic interventions or diagnostics into R&D is technologically challenging, and would benefit significantly from industry-wide pre-competitive collaboration.

“Pharmaceutical companies are capturing and storing more data than ever before. But deriving insights from data which translate into R&D outcomes that actually benefit patients, is a huge challenge,” commented Dr Steve Arlington, President, The Pistoia Alliance. “At the same time, today’s healthcare environments demand that pharmaceutical companies prove their therapies work, and that their cost can be justified. These goals can only be achieved if the industry collaborates to build cutting edge data analysis solutions, including deep learning and machine learning systems. Pharmaceutical companies cannot go it alone – unique solutions that are not interoperable or cannot share data are a considerable waste of time and money, which benefit neither patients nor payers in the slightest.”

A key event on the conference agenda was an update on The Pistoia Alliance’s President’s Series Hackathon. The hackathon was a series of five challenges, held over the 25^th-26^th March 2017, designed to bring the deep and machine learning community together with the life science and healthcare industries, to demonstrate the potential of deep learning to aid drug discovery and bring life-saving treatments to the world. The first challenge was delivered by Elsevier, in conjunction with UK-based charity, Findacure; it sought to accelerate treatment and clinical research for Friedreich’s ataxia (FRDA). The second challenge – the compound prediction challenge – was sponsored by ExCAPE, in conjunction with Janssen and Imec; the hacketeers were tasked with proposing innovative and performant predictive machine learning models for a number of assays.

A third challenge was on the ability of machine and deep learning to gain insights from social media forums into patient experience of a particular disease, such as asthma. The fourth challenge, sponsored by Promeditec, aimed to accelerate early diagnosis of Thoracic Aortic Aneurysm (TAA) through machine learning. Finally, a fifth challenge to predict potential disease-causing DNA mutations from the ClinVar public resource and Ensembl genome browser, was delivered by Microsoft. All of the five hackathon challenges were supported by Microsoft; which provided access to, and support for, its Azure cloud suite, Azure Machine Learning Studio, and Azure notebooks for machine learning and scripting.

The Pistoia Alliance’s European conference was attended by 120 life science professionals, representatives from top 10 pharma companies, leading biotechs and academic medical centres. Speakers and panel attendees converged on a range of topics; including patient data capture, deep data analysis, and ‘future thinking’ – such as, cutting edge therapeutics, companion diagnostics and reality mining. The Pistoia Alliance was created to address the barriers and challenges that companies face during life sciences R&D, and prevent the waste of time and resources resulting from siloed research programmes. Its projects bring together key constituents to identify root causes of R&D inefficiencies; it has also created a proven legal framework for open innovation. For more information about The Pistoia Alliance, please visit: pistoiaalliance.org.

The final sponsored challenge is here courtesy of the Pistoia Alliance member; Promeditec.
These challenges are designed to show how deep learning can work with life science and health related data to make an impact on advancing research into tackling disease and supporting patients.

The Challenge: Accelerate early diagnosis of Thoracic Aortic Aneurysm (TAA) through machine learning

Promeditec, as part of the under-development project InSilicoTrials.com, would like you to demonstrate the ability of deep learning to help accelerate early diagnosis and improve surgical treatment of Thoracic Aortic Aneurysm (TAA).
TAA is a “ballooning” of the upper aspect of the aorta, present primarily in the thorax above the diaphragm. The mortality rate as a consequence of untreated or unrecognized TAA, due to aortic dissection or rupture, was 2.78 per 100.000 in 2010, increasing from 2.49 per 100.000 in 1990.
Automatic reconstruction of the aortic walls, from patient’s medical images, combined with computational fluid dynamics simulations, based on patient-specific hemodynamics, can provide very important information (otherwise undetectable) on the evolution of TAA, to give surgeons precious indications during the surgical planning in order to reach the optimal patient-specific treatment.

Join us!
We look forward to seeing you at the event.
Register now.

For any questions on how to get involved as competitors, judges or provide a challenge, please contact David Proudlock.

We are delighted to announce that Microsoft is supporting our Deep Learning Hackathon.

The Pistoia Alliance want to bring the deep / machine learning community together with the life-science and healthcare industry to demonstrate the potential of deep learning to aid drug discovery and bring new life saving treatments to the world.

The hackathon aims to show how deep learning can work with life science and health related data to make an impact on advancing research into tackling disease and supporting patients.

Microsoft will help us achieve this by providing access to and support for their Azure cloud suite for the event and their

• Azure Machine Learning Studio
• Azure notebooks for machine learning and scripting

Join us!
We’ve already announced two of the challenges, more are being developed and will be announced shortly.
We look forward to seeing you at the event.
Register now.

For any questions on how to get involved as competitors, judges or provide a challenge, please contact David Proudlock.

We are delighted to announce the next of our partner challenges for the hackathon.
These challenges are designed to show how deep learning can work with life science and health related data to make an impact on advancing research into tackling disease and supporting patients.

The ExCAPE compound activity prediction challenge, presented by Janssen and Imec

Drug discovery comprises the identification, improvement and documentation of candidate drugs before their evaluation in patients; it involves the time-consuming and costly quantification of the activity of chemical compounds in various in vitro tests, known as assays. Machine learning approaches that can accurately predict the in vitro activity of compounds based on their chemical structure, promise to significantly reduce the time and cost to bring novel medicines to patients. On modest data scales and on an assay-per-assay basis, such approaches are already well established. But pharmaceutical companies are keen to also harness recent the advances in machine learning and big data analytics that are transforming other industries. They have the data to do that: the enterprise warehouse of a pharmaceutical company typically annotates millions of compounds with their activity in one or more of the many thousands of documented assays.
ExCAPE (Exascale Compound Activity Prediction Engine) is an EU-funded collaboration that unites pharmaceutical, technological and academic partners to harvest the power of supercomputers to speed drug discovery using machine learning. To enable cross-fertilization with machine learning advances in contexts other than drug discovery, the
ExCAPE partners have compiled and reformatted compound activity data from the public space for easy exploration. This dataset, which is a fair approximation of the compound activity data in a pharmaceutical data warehouse, is freely available to the field. As part of the Pistoia Alliance Deep Learning Hackathon, two of the ExCAPE partners, Janssen and Imec are formulating a challenge around this dataset.

A compound activity warehouse can be thought of as a table or matrix with compounds as rows, assays as columns, and in vitro observed activities as values. The fill rate of this matrix, i.e. the number of compound:assay combinations with an annotated activity is typically lower than 1%. For all the compounds, a sparse but extensive vector of binary features is provided that represents the chemical structure of the compounds.

The Pistoia hacketeers are challenged to propose innovative and performant predictive models for a handful of assays (columns) that will be disclosed at the start of the event. The assays to model are selected to be expected to benefit from the inclusion of activity data in other assays in the datasets during model training.

Join us!
More challenges are being developed and will be announced shortly.
We look forward to seeing you at the event.
Register now.

For any questions on how to get involved as competitors, judges or provide a challenge, please contact David Proudlock.

We are delighted to announce the first partner challenge for our President’s Series Hackathon.
The challenges are designed to show how deep learning can work with life science and health related data to make an impact on advancing research into tackling disease and supporting patients.

Elsevier would like you to demonstrate the ability of deep learning to help Findacure, a UK-based charity, accelerate treatment and clinical research for Friedreich’s ataxia (FRDA). You’ll have access to a heterogeneous set of data related to the disease: biological pathway analysis, associated chemical compounds and bioactivities, potential candidates for drug re-purposing, full-text scientific literature, and clinical trial data.

About Friedreich’s Ataxia
Friedreich’s ataxia is a rare, genetic, progressive, neurodegenerative movement disorder, with a mean age of onset between 10 and 15 years. Rare diseases are often assumed to be unimportant because of their small patient populations. This leads to a lack of support for patients and families, neglect by the medical profession, and little ongoing research into treatments.

Join us!
The aim is to forge new relationships between science and deep learning experts to look at established healthcare data in a different way. So if you are a scientist or a deep / machine learning specialist we want you to be there. Other disciplines are welcome, if you have a creative mind and this topic interests you please consider joining.

More challenges are being developed and will be announced shortly.
We look forward to seeing you at the event.
Register now.

For any questions on how to get involved as a competitor, judge or provide a challenge, please contact David Proudlock.