Bringing Artificial Intelligence to Cardiovascular Medicine and Cancer: Genomics in Action

WuXi NextCODE Nature

A Yale research team, with contributions from WuXi NextCODE’s artificial intelligence (AI) and sequencing teams, has discovered a novel mechanism regulating how blood vessels grow.

Artificial Intelligence (AI) can already catch a criminal and identify the right patients for certain types of surgery. But those challenges involve relatively few parameters compared to number of parameters or features involved in linking the 3 billion bases in the human genome with other ‘omics data and all the complexity of human biology. For that very reason, the promise of AI in genomics is as necessary as it is enticing, and WuXi NextCODE is committed to pushing the frontier of this emerging field.

This week, I am encouraged by results from a study published in the latest edition of Nature, which describes how a well-known growth factor may play a previously unknown role in some important diseases. That report, led by Yale University scientist Michael Simons, investigates blood vessel growth regulation—a crucial process in some very common conditions, including cardiovascular disease and cancer. Our Shanghai team provided RNA sequencing for this study. Our Cambridge AI team applied some of the most advanced statistic in their toolset to take the data analysis to the next level.

Simons’ team studied knock out mice, whose fibroblast growth factor (FGF) receptor genes were turned off. The scientists were able to prove, for the first time, that FGFs have a key role in blood vessel growth, uncovering some metabolic processes that were “a complete surprise,” according to scientists on the team. Further, they mapped out pathways that could help provide new drug leads.

It’s inspiring to see scientists from around the world using top-notch technology to collaborate on pivotal research questions. This study involved scientists in six different countries.

This FGF study also comes on the heels of our recent announcement about how our deepCODE approach classified 27 different tumor types with greater than 95% accuracy when applied across approximately 9,000 human tumors from The Cancer Genome Atlas (TCGA) collection. [LINK: https://www.wuxinextcode.com/highlights/posters-at-aacr/#/brief–using-ai]

With the rapid rate of progress, it’s not surprising that AI is finding success in genomics. Today’s informatics capabilities allow for assimilating larger and larger datasets with AI applications, and the field is evolving at a rapid pace. Google alone published more than 200 papers on AI in 2016. Like us, they use a deep learning approach.

From facial recognition to genomic solutions
Each AI problem has a different scale. In facial recognition, AI applications analyze relatively few features in the human face (about two dozen). Digital scans of the human eye that use AI techniques are able to segment patients before eye surgeries, and this entails algorithms that consider hundreds of features.

Genomics, of course, involves looking at any number of feature sets among billions of possibilities. It’s an immense challenge, but I think it’s perfectly suited to AI. And with deep-learning tools, we can fish out many more insights than with traditional analyses.

Our goal is to see how AI can help researchers achieve better results in identifying and evaluating new medicines, pinpointing risk factors and disease drivers, finding new combination therapies that work better than single drugs, and more. Our deepCODE tools comprise a novel, multinomial statistical-learning method and deep learning classification strategy. It’s an advanced approach to AI.

This week’s Nature paper is another encouraging sign.

Many of the stickiest problems in medicine are longstanding. The role of FGFs in blood vessel development was poorly understood until now. This group’s findings may help open new avenues of research.

Our team is always seeking to tackle problems with the latest approaches and technologies. Now, in the age of big data, it makes sense to start letting computers do more and more of the work, even some of the actual thinking. Certainly, we pick the questions and frame them. But then, let’s load the data and let the machines help us find the answers. If we can polish this process, and apply it to a growing number of problems, new answers and insights are sure to come.

email

Marking Progress in Genomics: Reflections and Prospects

Progress in Genomics WuXi NextCODEAs leaders of our field gather in Vancouver for the annual American Society of Human Genetics Meeting (ASHG 2016), it is an excellent time to take stock of the past and clarify our perspectives for the future. For the field of genomics, this is an opportunity both to reflect on our accomplishments over the last few years and to consider what we can achieve in the years ahead.

Indeed, our accomplishments have been numerous and our goals are ambitious, yet achievable. Here, I would like to summarize five significant ways in which our work in genomics has been revolutionizing medicine and improving patient outcomes.  In addition, I would like to share my thoughts about five areas in which I believe our field can drive meaningful change over the next few years.

What We Have Achieved
1. Improvements in Sequencing Technology and Analytical Tools
The ever-increasing volume of genomic data is testimony to the dramatic increases in sequencing speed and efficiency over recent years.  At the same time, novel methods of analysis, like the powerful genomics platform employed here at WuXi NextCODE, have considerably advanced our understanding of genetic variations and their clinical significance.

2. Transformations in Cancer Treatment
As I have discussed here, the expanding use of genomic data to guide treatment decisions in oncology is transforming the way clinicians approach cancer treatment.  In addition, our growing understanding of genetic predispositions for certain cancers is helping high-risk individuals make informed choices about preventive care.

3. Progress in Rare Diseases
Genomic data has brought new hope to families struggling with rare diseases by shortening diagnostic odysseys, guiding treatment, and building communities.  I provide examples of the game-changing power of genomics in the diagnosis of rare diseases here.

4. Empowerment of Patients and Consumers
Patients and consumers are increasingly informed about the innovative and meaningful ways in which genomic data can guide healthcare decisions.  The successes in our field are empowering individuals to pursue personalized medicine and generating interest in direct-to-consumer testing.  I offer my thoughts about DTC genetic testing here.

5. Innovations in Cloud-Based Analysis
The vast and ever-growing quantity of genomic data and related information necessitates new approaches to storage and analysis.  As I have previously discussed, cloud-based computing is essential to continued success in genomics.  WuXi NextCODE’s Exchange is at the forefront of the accelerated research made possible by real-time collaboration and analysis in the cloud.

What We Can Achieve in the Years Ahead

1. Effective Communication and Collaboration
Realizing the full potential of big data and cloud-based computing will require new efforts to dismantle “data silos.”  I am encouraged by recent initiatives to facilitate collaboration in cancer research, and – as I have recently discussed – call upon researchers and clinicians throughout the field of genomics to improve communication among all stakeholders.

2. Policies for Research with Patient Data
Our field derives its greatest power from careful analysis of genomic data, and access to data is critical to effecting meaningful change in healthcare.  In order to gather this game-changing data – from patients, from consumers, and from population-wide studies – we need to develop and embrace policies that lead to consider the ‘biorights’ of patients.  Individuals who wish to contribute information for research should have the opportunity to do so, and all parties should clearly communicate the purposes and extent of data-sharing.

3. Integration for Clinical Trials
I perceive significant movement toward the development of clinical trials that test the efficacy of treatments tailored to specific genetic anomalies – and use genetic information to screen participants.  This is an area in which genomics will dramatically accelerate the development of personalized therapies that will surely improve patient outcomes.

4. Actionable Information from Population-Wide Genomic Studies
I believe that in the near future we will reap significant rewards from projects that gather population-wide genomic information.  Analysis of the data we are collecting around the world, which I describe here, is an essential step to reshaping healthcare practices worldwide.

5. Globalization of Genomic Products: ‘Think Globally, Act Locally’
The power of genomic information is now known throughout the globe, and can be applied in a multitude of positive ways.  With such widespread potential, individual countries and cultures will choose to advance and roll-out genomics in their own distinct ways for the benefit of their citizens.  Companies that develop genomic products will need to adapt and design their products for use in specific markets.  At WuXi NextCODE, the first focus of our product portfolio for individual patients and families is in China, where we are delivering three offerings: population-optimized diagnostics, carrier screening, and whole-genomic wellness scans.

Together these initiatives build upon our recent accomplishments and further the creation of data and analysis necessary for meaningful change in healthcare.

The genomic revolution in medicine that we envisage will be achieved through applied use of research and development that is:

  • Fueled by big data, including data provided by informed consumers and patients and data derived from population-wide studies;
  • Supported by clinical trials crafted to assess the safety and efficacy of treatments tailored to individual characteristics; and
  • Enabled by collaborative work and effective communication.

At WuXi NextCODE, we are energized by the prospects for genomics in the years to come. We are proud to be at the cutting edge, providing the tools and resources that researchers and clinicians need to harness the transformative power of genomic data. And—importantly—we are confident that our field will continue to drive meaningful changes in healthcare that improve patient outcomes.

Genomics: Big Data Leading to Big Opportunities

The Big Data of Genomics

WuXi NextCODE Exchange

The big data of genomics will continue to expand, and our approaches to analyzing genomic data need to continue to evolve to meet the growing demands of clinicians and researchers. Cloud-based platforms such as WuXi NextCODE’s Exchange are essential to address the fundamental big data challenge of genomics.

Beyond question, we are in the midst of an explosion of “Big Data” in many facets of human endeavors. In fact, data-storage leader IBM asserts that roughly 2.5 quintillion bytes of data are generated every day and 90% of the world’s data was created in the last two years.

An outpouring of articles in scientific journals and major newspapers has highlighted the promising potential of big data in medicine, including a special section in the current issue of Nature.  Genomics has become a major source of the growth of such big data, particularly as the cost of sequencing genomes has plummeted. The raw sequence data for just one person’s whole genome use as much as 100GB—and already hundreds of thousands of individual genomes have been sequenced.  With more than 2,500 high-throughput sequencing instruments currently used in 55 countries across the globe, more genomes are added every day. The aggregate amount of genomic data is growing explosively, and next-generation sequencing (NGS) sequencing data are estimated to have doubled in volume annually since 2007.

The accumulation of genomic data is a worldwide phenomenon.  Impressive population-wide sequencing efforts are leading the way, from 100,000 genomes in England, Saudi Arabia, and Iceland to 350,000 in Qatar to a million in both China and the U.S.

And earlier this month, the CEO of the Cleveland Clinic predicted that soon children will routinely have their whole genomes sequenced at birth, implying a near-future in which 10s of millions of new genomes are sequenced annually.

Turning Data into Resources

But sequencing genomes is not enough, and the creation of genomic big data is just the beginning.

Thanks to the analysis of big data in genomics and associated informatics, we are seeing meaningful progress in cancer care and the diagnosis of rare diseases, as I have discussed here and here. We clearly have a tremendous opportunity to use the big data of genomics to continue to drive a revolution in healthcare.

Yet there is a broad consensus that a ‘data bottleneck’ is hampering collaboration and discovery. Not all researchers and physicians confronting the current onslaught of genomic big data can readily determine how to use genetic information to prevent or treat disease. To succeed, researchers and physicians clearly need resources that:

  • Draw together useful data from disparate sources;
  • Facilitate analysis and collaboration; and
  • Improve clinical practice.

The power of genomic analysis needs to expand outward from major research centers and hospitals to the myriad clinics and community hospitals where many patients receive care. To have the greatest impact on the broadest population, clinicians throughout the world’s health systems need access to the big data generated by DNA sequencing, even—or perhaps especially—if they are not affiliated with research institutions. They also need to be able to make sense of the data they have access to.

Answers in the Cloud

Sequencing provides the raw data to uncover the genetic variants that contribute to disease. But the datasets are too big to transfer repeatedly—and too big even for smaller hospitals, labs, or clinics to store onsite. Key medical advancements require not only big data, but also tools and resources to generate, interpret, and share analysis of millions of genomes.

Cloud-based platforms—such as WuXi NextCODE’s Exchange—are essential to address the fundamental big data challenge of genomics. Collaboration in the cloud works to dismantle existing “data silos”—genomic information hosted only on local servers and analyzed on idiosyncratic, closed platforms. The NextCODE Exchange, in contrast, is a browser-based hub that affords secure, seamless collaboration with colleagues around the world. Moreover, users get access to NextCODE’s tools for making the critical links between variation in the genome and disease and other phenotypes, backed by harmonized links to the the most important public reference data.

And cloud-based computing is inherently scalable: resources for data storage and analysis expand as needed, allowing researchers and physicians to leverage massive datasets to improve patient care in the clinic. The big data of genomics will continue to expand, and our approaches to analyzing genomic data need to continue to evolve to meet the growing demands of clinicians and researchers.

At WuXi NextCODE, we have built upon our heritage of conducting the largest analysis of genomic data (deCODE’s path-breaking Icelandic analysis) by assembling an ever-growing database of human genomes. We are committed to driving the movement of sequence data into patient diagnosis and care through user-friendly, leading-edge analysis and informatics. I am confident that data analysis and collaboration in the cloud will revolutionize healthcare, and exceptionally proud that WuXi NextCODE’s Exchange is at the forefront of this exciting advancement.

Genomics in Cancer: Continuing to Push the Leading Edge

genomics in cancer - hannes smarason

Genomics is helping to prevent and treat cancer at an accelerating rate, supporting the goal of oncologists to dramatically improve cancer patient outcomes.

The progress in the use of genomics to help prevent and treat cancer continues to grow at a pace that is impressive. Indeed, there is expanded use of genomics to drive patient care and improve outcomes across an ever-expanding number of cancers by a growing number of oncologists.

Genomic Knowledge Can Clearly Drive Better Care

Applying genomics to cancer treatment is a powerful clinical application, as genomics can provide a window into how to best treat a patient’s particular cancer as it:

  1. may help better understand the genetics of the tumor itself, and
  2. can provide insight into how cancerous tumors may grow and spread over time.

With a genomic-based approach to cancer care, oncologists can more personally tailor anti-cancer treatments to an individual tumor’s mutations, thus molecularly targeting the specific cancer’s Achilles heel. Already, there are well-documented successes of molecularly targeted anti-cancer agents, such as cancer drugs that target certain genes—HER2, EFGR, ALK, and others.

In 2015, the pace of adoption of genomics in clinical oncology has advanced significantly. Recent evidence of the accelerating use of genomics to help fight cancer includes:

  • Evolving from ‘why’ to ‘how’ to use genomics at leading cancer centers. At the top cancer care facilities, genomics has become part of the programmatic approach to provide certain cancer patients with optimal care—care that is fundamentally designed to lead to the best outcomes. The question for leading medical centers globally has evolved over the last few years from “do we need genomics?” to “for which cancer types and at what stages of cancer treatment and diagnosis can we best use genomic sequencing and analysis?”—an evolution from “why?” to “how?” at a very fundamental level. The accelerating use and deployment of genomics by leading medical facilities validates that they are deriving significant value from genomics, and that value is resulting ultimately in meaningfully advancing better care for cancer patients.
  • Expanding potential applications of genomics within different types of cancers, broadening the types of cancers and tumors that can potentially benefit from genomics. Researchers and clinicians continue to publish a wealth of information validating the potential of genomics to improve outcomes in certain types of cancer patients. In 2015 alone, highlights of these advancements include certain prostate cancers, brain cancers, rare types of pediatric kidney cancers, and even potential targets in certain non-small cell lung cancers.
  • Broadening acceptance in cancer prevention. Driven in part by the education of oncologists and physicians generally and in part by the empowerment of knowledgeable patients, people are seeking and benefiting from genetic tests that reveal their personal risk for certain tumors (such as BRCA for breast or ovarian cancers). The idea of using genomic analysis to predict an individual’s cancer risk by comparing their genome with databases of confirmed genetic mutations linked to disease is—for certain individuals with specific family histories and genetics—driving appropriate medical decisions for patients who may be at high risk for certain cancers.
  • Powering clinical trials with genomics. The use of genomics in cancer clinical trials – whether for inclusion in data-gathering or even screening of patients—has gone from rare to commonplace over recent years, and is improving knowledge around the safety and efficacy of drugs in cancer and beyond. Two large-scale cancer trials have been initiated in 2015 with the bold goal of substantially advancing the understanding and use of genomics in cancer care. The anti-cancer treatments being tested in both trials were selected for their activity on a specific molecular target, independent of tumor location and histology. The two trials are actively enrolling and are (1) an American Society of Clinical Oncology (ASCO)-sponsored study, called TAPUR (Targeted Agent and Profiling Utilization Registry) and National Cancer Institute (NCI) and is called NCI-MATCH (Molecular Analysis for Therapy Choice). These trials and any subsequent follow-on trials will doubtless provide insightful information to drive the growing use of genomics in improving cancer care.

In summary, genomics is helping to prevent and treat cancer at an accelerating rate, supporting the goal of oncologists to dramatically improve cancer patient outcomes. There are at least four frontiers where we can see substantial progress in the use of genomics in cancer care, including expanded use in leading medical centers, increased potential applications within cancer, widespread acceptance in cancer prevention, and an increase in the use of genomics within clinical trials. I am personally committed to continue to drive and accelerate this genomic revolution to continue to bring true progress in improving cancer care to patients in need globally.

Maintaining Momentum Post-ASHG: Maximizing the Value of Large Genomic Databases

The newly launched NextCODE Exchange provides a browser-based hub for multi-center sharing and collaboration on collective data from massive whole-genome databases like the Haplotype Reference Consortium (HRC).

The newly launched NextCODE Exchange provides a browser-based hub for multi-center sharing and collaboration on collective data from massive whole-genome databases like the Haplotype Reference Consortium (HRC).

The American Society of Human Genetics (ASHG) meeting convened this week in San Diego, bringing together genetics experts from around the world to discuss programs with great potential to advance genomic-based medicine in the years to come.

To maintain the momentum generated this week, we need to find ways to integrate these important ideas, insights and programs, and to maximize the use of the massive databases that have been launched to support research on cancer, rare diseases and other pressing health topics.

One of the databases unveiled during the meeting was the Haplotype Reference Consortium, which aims to become the world’s most comprehensive database of genetic variations. Large databases like the HRC, along with several others already underway, can be tremendously helpful to researchers finding answers to some of the most challenging diseases. But there remains a significant bottleneck: these large, cumbersome databases cannot easily be shared and manipulated, limiting their utility for broad, multi-center genomic research.

The solution lies in the newly launched NextCODE Exchange (see release here). This browser-based hub allows for the sharing and harmonizing of massive whole-genome databases like the HRC to accelerate research. The integrated architecture allows users to visually confirm and validate findings in raw sequences, collaborating and sharing with others around the world who may have complementary research underway.

The momentum generated during ASHG will be multiplied by sharing and learning from the world’s collective genomic data on the NextCODE Exchange. Learn more here.

Pioneering Genome Sequencing Effort in England Aims to Shape the Future of Global Medicine

£300 million in new investments for Genomics England

Genomics England 100,000 Genomes Project

Genomics England was set up by the UK Department of Health to deliver the 100,000 Genomes Project. Initially the focus will be on rare disease, cancer, and infectious disease. The project is currently in its pilot phase and will be completed by the end of 2017.

These are exciting times for large-scale sequencing projects. Last week, U.K. Prime Minister David Cameron announced over £300 million ($509.4 million) in new investments for Genomics England, which aims to sequence, analyze, and store the genomes of 100,000 UK National Health Service (NHS) patients by 2017. The investments include about £162 million ($275.1 million) from Illumina Inc. (NASDAQ:ILMN), the partner for the sequencing element of the project. In turn, Genomics England will pay Illumina about £78 million ($132.4 million) for its services.

At the same time, the Wellcome Trust will put £27 million ($45.8 million) into a new sequencing hub at its genome campus in Cambridge; the Medical Research Council, or MRC, is investing £24 million ($40.7 million) to support data analysis and interpretation, and the NHS will make £20 million ($34 million) available for the establishment of patient sequencing centers.

This is a prime example of how the implementation of sequencing technologies promises to drive a revolution in the structure of medical research. These new projects aim to capture more data on human DNA than ever before, with the goal of advancing care and solving healthcare challenges.

The 100,000 Genomes Project, developed by the NHS, has the potential to significantly influence the global community through its plans to integrate sequencing data into standard medical practice.

Genomics England plans to generate 100,000 whole genome sequences from NHS patients with cancer, rare diseases, and other conditions, and to share the resulting data for research and development purposes. In the early phases, the program will also seek to develop standards for consent, sample storage, data generation and variant analysis that may be useful for many other organizations conducting large-scale projects within public health systems.

The project is enlisting the help of organizations from around the world to undertake this significant effort. In fact, it recently selected Illumina to conduct the sequencing efforts and is evaluating technologies for storing, annotating, and interpreting the data so that it can be used  for both clinical diagnostics and drug discovery, development, and delivery to the right patients.

The challenges of analyzing data on such a large scale are formidable, but the end result carries great potential to address some of the significant unmet medical needs. NextCODE’s technology has already accomplished analytics on this scale based on its work with the Icelandic population through deCODE genetics. It’s an exciting prospect for advancing the future of genomics-driven medicine and one to watch.