Pharma: Trends and Uses Of Mobile Apps and Digital Marketing

A new study suggests that printing drugs on a QR code that patients can scan with their phones could pave the way for personalized medicine.

Via Rémy TESTON, Lionel Reichardt / le Pharmageek

After five years of development, Janssen is finally set to launch the first clinical trial to use its iSTEP technology toolset, which employs smart blister packs and a patient-facing smartphone app.

Via Dominique Godefroy, Lionel Reichardt / le Pharmageek, Philippe Marchal

Running clinical trials is an expensive business and Novartis has signed a new deal with US tech company Science 37 to try to take some of the costs out of the process.

The aim is to expand the number of ‘virtual’ trials it runs, with patients interacting with investigators via mobile phones and telemedicine devices rather than attending investigation sites in person. It’s not a new concept - Novartis already has virtual trials ongoing - but the deal with Science 37 for ten trials over the next three years signals greater acceptance of the approach.

It’s a gradual process, and Novartis says the studies will “blend virtual and traditional models, with increasing degrees of decentralisation towards a mostly ‘site-less’ model”. This could allow trials to dramatically increase the pool of patients they can draw upon as they can now participate at their home or via their own doctor’s office.

Under former R&D chief and now CEO Vas Narasimhan, Novartis has long been a devotee of using technology to help the clinical trials process, medical education and healthcare delivery become more efficient.

In 2015 for example the company formed a wide-ranging alliance with tech giant Qualcomm to move towards greater use of digital devices in clinical trials, and it has also been active in making venture capital available to players in this sector - in fact it took a 10% stake in Science 37 after participating in the firm’s third-round financing last year.

Novartis says the new decentralised trials are expected to begin later this year in the US in the areas of dermatology, neuroscience and oncology. It’s already running semi-virtual trials for non-alcoholic steatohepatitis (NASH), acne and cluster headache.

Via Dominique Godefroy, Lionel Reichardt / le Pharmageek

Your digital marketing challenges are probably shared by peers in other departments and companies. The good news is that such common problems ofte

UnitedHealth Group has partnered with DexCom to launch a glucose management program using a wearable device to help older Americans manage their Type 2 diabetes via mobile technology.

What's the role of social media influencers in healthcare and pharma? We've got the details for you here along with seven engagement models.

Via Lionel Reichardt / le Pharmageek, Philippe Marchal

Prior to booking an appointment:

 

• 77% of patients used search

• 83% used hospital websites

• 54% used health insurance company sites

• 50% used health information sites

• 26% used consumer-generated reviews

( Source: Google’s Report on The Digital Journey to Wellness: Hospital Selection )

Digital Marketing should now be part of your overall marketing strategy, no matter which industry you are in. As you might have guessed from above, it is true for Pharma Industry as well. Digital marketing in pharma industry has already picked up and starting to transform the Pharma and Healthcare industry in the ways it has already transformed retail, media, banking, airline, telecom and education industries.

Pharma companies and its executives have realized the potential and the disruption it can bring-in the pharma sector and have already started making use of and experimenting with the wide range of armaments available in Digital Marketing.

Trends that make Digital Marketing Imperative for Pharma Companies

Today’s generation turns to internet and social media for almost everything in their lives, health topics would be no surprise. However, there are certain trends which are acting as driving force to make Digital Marketing for Pharma companies a must-have.

Let’s look at such key trends:

 

• Patients are becoming more aware, more engaged and have more expectations

Historically, the patients have had a much passive role when it comes to their own health treatments. They would simply go and meet the doctor and rely on the medicines prescribed by her and start the treatment. This is because they have very little to no information about the treatment options, drugs on the market and the experience of other patients.

With digitization, before meeting a doctor, the patient researches everything about the disease – its symptoms, diagnosis, possible cures etc and has already interacted with other patients. Patients are also more and more aware of their rights and have high expectations from the service providers and the product companies alike.

 

• More information is available about product performance and even the process

Traditionally, pharma companies controlled all the information regarding their products – and they would also control release of this information. They usually used to release this information on need basis (e.g. required by regulators).

The digitization has weakened the control that pharma companies had on information. There is abundance of information available online – people search online for solutions to all of their daily problems. Health problems are no exceptions. Patients share their experiences with drugs, doctors and companies online, which are available for other patients to see. Therefore, in this digital age, patients are increasingly less dependant on their doctors for advices. Instead they rely on plethora of information available online, on mobile apps and they use monitoring devices (e.g. fitbit) and monitoring apps too.

“Two-thirds of people believe they could be making more decisions about personal health and wellness on their own.”

– A survey conducted by Ipsos in collaboration with the National Council on Patient Information and Education and Pfizer, 2015, multivu.com.

This number would only have gone up since 2015. Patients are increasingly turning to internet and social media not for supplemental information to add to what they got from doctors and pharma companies but as first source of information.

 

• Process efficiency is improving exponentially

Process efficiency is going up across industries with data analytics and automation of complex decisions, in turn improving agility, responsiveness, accuracy and quality of business processes. Pharma industry is no exception, they would also need to apply next-generation technologies to stay in the game.

 

• Tailored personalized care

With increased expectations from the whole eco-system, users not only want but rightfully expect personalized care. They are able to get personalized solutions – be it grocery shopping, electronics shopping, spa – salons, banking services, online trading services, telecom services, airline companies and a lot more.

Users expect the same from pharma companies as well. “Patient First” has become the war cry for pharma and healthcare companies, in line with Customer First for other traditional sectors.

Pharma companies can provide personalized care through the right usage of digital services, sensors, tech-enabled devices and through processing the data collected by these. This makes a really strong case for digital marketing in pharma industry.

 

 

 

• Multi-channel engagement between patients and doctors

The patients are already starting to use online portals to access their medical records and to interact with their doctors. Patients are starting to use apps to fill forms online and engage with other patients in the online communities.

All of this opens up a whole new set of mediums for pharma and healthcare companies to interact with patients and other users. Using these mediums, the pharma and healthcare companies, through their sales representatives, patient-services teams and other teams, can monitor and influence patients, doctors and other healthcare professionals online through apps, social media, mobile phone messages/notifications and even in person.

Going forward, 24/7 anytime anywhere virtual healthcare is going to become a norm. Leading indicators of such norms can be seen by the developments in the early adopters of technologies.

The US Department of Defense is testing robots to engage and screen soldiers for post-traumatic stress disorder (PTSD) –

“The military is building brain chips to treat PTSD,” Defense One, May 28, 2014, defenseone.com.

United kingdom has ambitious plans to halve the number of outpatient appointments with use of skype consultations. Follow-up advice and prescription would be sent by email.

We believe that these futuristic steps may face some teething problems, but these are definitely the way of times to come.

Image: Digital Marketing in Pharma and Healthcare Sector

 

• Don’t just provide drugs, provide full service and better sales practices

A pharma or a healthcare company, like any other profit making business, would want their customers to come back to them whenever they need these services. The companies have to go the extra mile and provide full service support – before (decision making process), during and after sale of product/ service. The idea is to engage the customer during the full process and keep her engaged afterwards, like a personalized healthcare solution provider. Digital marketing is poised to play key role in this arena for the pharma and healthcare sector.

 

• Timely and accurate feedback for R&D

Use of digital technologies provides real-time feedback from the whole value-chain including the end-consumer to pharma and healthcare companies. It cannot be stressed enough, how important is that to their research & development centers. In effect, what I am saying is, digital marketing is going to help pharma companies in their R&D, and in turn will lead to better products and solutions for all of us. Also, it is likely to make the R&D expenditure for the pharma companies much more efficient.

And the seamless information flow will help the supply chain management better as well (remember the beer game?). The pharma companies can better predict demand and supply, security of their drugs and vendor/partner performance.

 

• Better management of Risk and Compliance functions: Monitor risks and better allocation of resources to manage the same in real-time

The real time feedback from the whole supply chain and end-users, provides the pharma and healthcare companies the capabilities to monitor risks in real time. This in turn enables them to respond in real time by optimizing resource allocation to different activities.

 

• Tech-enabled Competitors are Marching in

There was a time when information and insights into clinical pathways and patient’s histories were available only in the traditional healthcare establishment – in the paper records of healthcare providers. These used to remain in the clutches of pharma companies.

Today, the big technology giants such as IBM, Apple and many new startups are making in-roads into the healthcare segment. These new players engage with the patient through digital mediums – apps, fitness and health devices, online forums, video channels, live sessions etc.

They continuously collect data from their users, such as health records, drug purchases, health insurance premiums and claims etc. Then they use advanced data analytics to deliver personalized and precise medical assistance.

Pharma companies will have to tackle the competition arising from these new players. They will either have to create these capabilities themselves or collaborate with some of the new players, or think out-of-box to maintain the leadership position.

Thus, both the customers and the healthcare value chain partners as a whole are moving towards digital – finding solutions/ options online, engaging with other patients/ doctors online. The traditional medical seminars, conferences, magazines and meet-up are evolving too with changing times.

Digital marketing is going to provide the edge required to stay ahead in the race, for pharma and healthcare companies.

Via Plus91, Giuseppe Fattori

The pharmaceutical and healthcare industry is rapidly coming to terms with digital technology.

New tech companies are entering the market and existing players have had to up their game, across marketing, sales and operations. If you're relatively new to pharma, like me, here's a summary of just some of the biggest trends.

1. Better patient communication

A fairly obvious area for improvement, and one we can pretty much all relate to as consumers.

The typical patient journey is ripe for digital disruption, as the diagram below from DRG Digital shows. From the healthcare provider's point of view, the visit, diagnosis, treatment selection and condition management stages are all points where the patient could be more involved / better updated.

Patient portals, apps and online communities are increasingly commonplace. The second and third generations of this technology should help improve customer experience.

Click to enlarge

2. Providing services, not just drugs

It's fairly obvious that whilst drugs are vital in treating many conditions and diseases, there is much more for the patient and physician to consider. That could be anything from education or lifestyle advice, to emotional support. Pharma companies have always engaged with the end consumer but digital technology ultimately promises much greater scale.

One such example is AstraZeneca's Day-by-Day coaching service for patients recovering from a heart attack. The service provides a combination of digital content and one-to-one coaching, partnering with the HIPAA-compliant Vida app.

The app helps patients to manage stress, find information, learn about caregiving and chat with a coach. Novartis has taken a similar approach with its Together in HF social network.

The ultimate aim, of course, is for pharma companies to look more to outcome-based solutions which involve greater engagement with patients and third parties. In order to achieve this they must not only assist with holistic patient care in this way, but must also advocate a combination of therapeutics, whether or not they themselves manufacture them.

Providing this type of customer experience may be quite a step change, but pharma has for a decade been coming to terms with the internet's impact on patient knowledge and behaviour.

Consumers are ever more motivated by finding the best treatment and the cheapest price, and pharma must provide the best outcome-based approach to cut through. Indeed, some companies are partnering with third-party tech companies in an effort to promote an unbiased presence in the market.

3. Improving diagnostics and adherence

The concept of the connected human being given personalised care and improved diagnostics is one long foreseen by science fiction. Since wearables entered the market, the idea of ongoing measurement has seemed less fanciful. Indeed, the ubiquity of the smartphone now gives the prospect of reliable access to patient data in the real world. 

One well-publicised example of digital medicine is the ingestible sensor in development by Proteus Digital Health. The company received FDA approval in 2012 for a drug-sensor-app system, which is used to montor adherence.

When the patient takes their pill, it dissolves in the stomach and causes a small voltage (as a small amount of magnesium and copper come together). This voltage is then picked up by a sensor on the body (stuck to the arm) which relays the information (time of ingestion) to a smartphone app and then on to the physician.

Proteus has trialled this system with an antipsychotic and a hypertension pill. The obvious extrapolation is to a future where a range of patient data is securely transmitted to the physician and this will mean less time spent on diagnostics (or more accurate diagnostics) and more on personalising treatment.

This aggregation and analysis of patient data is a field where the seeds of disruption in pharma could already be sown. Watson Health Cloud was created in 2015, in partnership with Apple, Johnson & Johnson and Medtronic, and provides analytics services to healthcare professionals.

Watson Health Cloud analyses data from personal devices, connected medical devices, implants and other sensors, and can support clinical decisions and reduce incorrect diagnoses. Consumers may be motivated to use services such as Apple HealthKit as they pursue value and the best treatment.

This is an area where pharma companies will have to keep step, especially when it comes to proving the value of their drugs. Payors, too, may use more patient data to determine payment.

However, it's worth pointing out that a big wave of extra data doesn't necessarily represent a panacea. The fully quantified self has many ethical and privacy concerns - as Watson builds out knowledge of genomics, clinical and exogenous data, these issues will continue to be debated.

4. Better sales practices

Back to some good old-fashioned marketing now. Too often, private healthcare professionals or providers must meet with multiple reps from the same pharma company. This is understandable given the expertise needed within each specialism, but from a customer point of view, busy healthcare professionals (HCPs) may be left wanting a more flexible solution.

Increasingly, pharma companies are using digital technology (both customer facing and back-of-house) to provide this. CRM systems can achieve a single customer view and digital communication channels can provide access to samples and resources (for HCPs and patients).

SKURA is one tech company that provides such digital sales enablement for life sciences, and it defines this role of supporting HCPs as 'the trusted concierge'. The aim is just like that of CRM in many other sectors, to 'deliver personalized messages to customers at the right time using the right channel in order to increase revenue and reduce costs.'

Digital sales aids and marketing are now firmly on the rep agenda.

A related area is that of more effectively targeting patients. Patient finder technology such as Vencore Health Analytics combines clinical knowledge with big data to allow drug manufacturers to identify potential patients that may have a disease that is hard to diagnose.

This is similar to the Watson solution discussed above, using health records, genomics and claim data. Such data analysis will not only ultimately improve treatment, but necessarily increase the number of diagnoses.

Havas Health partners with Vencore

5. R&D and supply chain efficiency

In a recent post on CRM, marketing automation and data management, I discussed the concept of XRM ('anything' relationship management). CRM is not just about increasing sales through customer-facing technology or through analytics. There's much to consider in drug production, too.

R&D can be improved by bringing real-time technology to bear on clinical trials, and the supply chain could benefit from better sales and operations planning. This would bring better productivity, inventory levels, and service levels.

A common theme in the digital transformation of any industry, the digitization of the supply chain represents a security risk to pharma but is a necessary step in meeting raised expectations from all parties.

6. Real-world data and drug development

The proliferation of health analytics solutions has implications for drug development, too. Manufacturers will have access to much more real-world data and this will undoubtedly assist in understanding the effects of a drug.

Dr. Amy Abernethy of Flatiron Health tells McKinsey, “I would want to know what adverse events there are before others surface this for me. With constant monitoring, you will find a lot of signals, and you will need to learn how to handle these signals with respect to reporting to the Food and Drug Administration. But this is not a reason to stick your head in the sand; this is how drug development is going to be done in the 21st century.”

Abernethy goes on to suggest that this means clinical informaticists must eventually rise to become business leaders in pharma, as they help companies come to terms with losing dominance of information about their products. 

So, not only will drug discovery increasingly be aided by digital technology (in predicting successful drugs), but so too will the monitoring of drug use on a wider scale. 

Via Plus91, Lionel Reichardt / le Pharmageek

How to get consistent international digital pharma marketing while meeting local affiliates’ needs.

Tech giant Amazon (AMZN) is in preliminary talks with generic drugmakers Mylan (MYL) and Novartis (NVS) unit Sandoz ahead of its potential entry into the pharmaceuticals space, according to a report from CNBC.

X
 
In after-hours trading on the stock market today, Mylan climbed 3.3%, near 37.80, after closing down nearly 1%, at 36.53. Novartis was flat after shares ended the day up 0.8%, at 85.80. Amazon lifted 0.4% after the closing bell.

Via Pharmacomptoir / Corinne Thuderoz, Giuseppe Fattori

Pharma has been accused of being on the dark side, but now one agency is encouraging it to go there. The dark social side, that is.

 

While the name “dark social” sounds sinister, it’s actually just a growing way that people share content on social media, according to Andrew Grojean, social media manager at Intouch Solutions: “Dark social is effectively word-of-mouth sharing. It’s when a user does things like copy and paste a URL and drop it into a messaging app, or copies a link on a website and send it in an email where referrals aren’t always passed.”

 

Dark social also includes other shared activity that can’t be tracked or analyzed, such as links or content sent in instant messages, texts, Snapchats, mobile apps or even emails.

 

The problem for marketers is that dark social activity accounts for up to 75% of all sharing on social channels, and that figure has been growing over the past few years. That’s quite a bit of unknown activity when it comes to a brand and its content.

 

Dark social is also likely a valuable share, Grojean said. People sending content links in private messages are likely to be sending to a close circle of family, friends and acquaintances. Studies have repeatedly shown that people trust those kinds of word-of-mouth endorsements from family and friends more than they do advertisements.

 

So how can a pharma marketer tell if they’re getting a lot of dark social traffic? Marketers in general can usually tell the difference between direct traffic from people who type a simple url into a browser and the pasted-in link that still shows up as direct traffic in analysis but seems unlikely to have been hand-typed because of a long and unwieldy URL.

 

The good news is that pharma can take some steps to track and improve dark social sharing information, and Grojean offered some suggestions.

 

First, companies can start tracking everything. They can add tracking URLs to sites, emails, and Facebook and other social posts by creating a unique URL for each so that when someone copies the URL into a private message or chat and the receiver opens it, marketers using analytics will be able to tell how the user got to them.

Via Pharma Guy, Giuseppe Fattori

With marketing a consistent issue for pharma, one which will likely continue to grow, getting personal is the only option for future success.

Via Lionel Reichardt / le Pharmageek

“The absence of pharma brands on social media creates a significant void of reputable healthcare information to aid patients.” posits Dawn Lacallade, LiveWorld. Why isn’t the pharmaceutical industry more active on social media? They would say advertising restrictions and other FDA regulations severely limit their ability to have a social media presence. There is a fear of discussing prescription medication in the uncontrolled environment of the internet. But the industry is missing a terrific opportunity to impact their entire constituency: patients, caregivers, employees, scientists and even their reputation.

Unmetric, a branded content analytics company, recently released a report that outlined social media trends for big pharma. They cited four silos where pharmaceutical companies are utilizing social media. All companies studied have excellent corporate social profiles. They are attractive and informative in a general way about the company, but they aren’t interactive. Most of the pharmaceutical companies have a career silo. It is interesting that the pharmaceutical industry has lagged other industries in setting up and managing an effective career site. No real clarity on why this has happened. There is little to no FDA regulation on advertising open positions.

About half of the pharmaceutical companies in Unmetric’s study have invested money and content into OTC brand profiles. Again these tend to be static/informative and not interactive. The biggest opportunity for big pharma is in the last silo defined by Unmetric, branded community properties. Patient’s have been and continue to turn to social media to research and understand their symptoms and diagnoses as well as trying to connect with other patients.

 

Under current FDA regulations it is hard for the pharma company to easily join the conversation to provide accurate, balanced info because regulations mandate that “within a single social post brands must provide accurate details on the benefits and risks associated with conditions and products.” Character limits and the speed with which interactions occur means a different approach is necessary. Pharma companies must talk about the disease rather than the product or drug itself. They must try to create a place where people gather who are concerned about one of these conditions. Trying to figure out what drives engagement and putting more effort and money into it will pay off for big pharma.

Social listening is another tool that biotech and big pharma under utilize. Gauging community sentiment about marketed drugs, learning about competitors and gaining insights to improve products, services and treatments are all achievable through social media research. Social media should be more of a pull than a push of information when done correctly. Kiran Mazumdar-Sahw, Chair and Managing Director of Biocon Limited, says, “Doctors clearly will drive this change, as will younger patients. The mindset today is still controlled by pre-internet key opinion leader doctors and older patients who are not tech savvy. As younger and tech-savvy doctors and patients populate our health care ecosystem, things will change and this change will occur rapidly after a certain inflection point which is not more than 3-5 years away. There will be an explosion of social media and mobile-based apps.” For the savvy biotech or pharmaceutical company it’s time to start investing in social media.

Via Plus91, Pharmacomptoir / Corinne Thuderoz, Lionel Reichardt / le Pharmageek

Pear Therapeutics has inked a deal with pharma company Novartis to work together on the development of two digital therapeutics for multiple sclerosis and schizophrenia. It’s the first time a pharma company has made a development deal with a digital therapeutic, according to Pear.

Via Florian Morandeau, Lionel Reichardt / le Pharmageek

Pharmaceutical, health insurance, medical and digital technology companies took brave steps in 2017 to strengthen their digital health market position with more or less success. Yet, last year’s business moves also suggest important trends: producing drugs alone without added digital services is not enough anymore for pharma; medical websites are becoming huge media outlets; and …

Blockchain startup Viant announced a new collaboration with pharmaceutical giant GlaxoSmithKline (GSK), Microsoft, and others to develop blockchain-based supply chain tracking solutions. Together they’re looking to accelerate the growth of Viant’s Ethereum-based blockchain supply chain platform in the pharmaceutical sector by using the distributed ledger technology to track intellectual property (IP) licenses as well as ensure that products are produced, transported, and stored in proper conditions.
Via Alex Butler, Bruno Demay, Lionel Reichardt / le Pharmageek

The return on investment in R&D by the top 12 companies in the pharma industry is at its lowest level for eight years, at just 3.2% of spending, says a new report.

That is a big reduction from the 10.1% return in 2010, made even more serious by a massive increase in the cost of bringing a new drug to market, now almost $2bn compared to $1.2bn in 2010 when the annual survey - by Deloitte’s Centre for Health Solutions - was first conducted.

Last year that figure was just over $1.5bn, and big increase in just one year is attributed to a decline in the number of drugs in the late-stage pipeline, which is down 16% on 2016 with 159 candidates across the industry.

One slightly bright spot in the data comes from Deloitte’s analysis of average peak sales per asset, which has risen from $394m last year to $465m but still lags well behind the $816m figure reported in 2010 and that overall decline contributes to the low internal rate of return.

The data is “a stark reminder that investing in R&D is a high-risk, high-reward endeavour”, according to Colin Terry, consulting partner for European life sciences R&D at the accounting organisation, who suggests that the industry is facing big challenges including “increased competition, expiring patents, declining profitability, mounting regulatory scrutiny, and arguably the most heated issue - pricing”.

Via Dominique Godefroy, Lionel Reichardt / le Pharmageek

The FDA just broke a recent record for most new drug approvals in a year, hitting a not-entirely meaningful (but not altogether meaningless) milestone for an agency that has long promised to pick up the pace.

 

With today’s green light of La Jolla Pharmaceutical Company’s Giapreza, a treatment for dangerously low blood pressure, the FDA hit 46 approvals for the year. That’s the most in at least a decade.

 

We’re looking at only approvals from the FDA’s drugs division, which considers pills and injections that treat disease. So-called living products, including vaccines and gene therapies, are approved by a different division at the agency and counted separately.

 

Further Reading:

• “Big Pharma Had a Bumper Crop of New Drugs Approved in 2017, But Profitability Shrinks”; http://sco.lt/7OtdVx

Via Pharma Guy, Lionel Reichardt / le Pharmageek

The pharmaceutical industry may have pioneered some great science, but when it comes to digital health it has been behind the curve. As patients and payers open up to the idea pharma has now begun to embrace the digital revolution. This report looks at where the industry is now and where it can go.

The swift developments in digital technology in recent years have transformed so many areas of our lives, from education and personal banking, to everyday activities like booking travel. Healthcare has also seen digital incursions, yet the strides made here seem small compared with those achieved in other areas – the motor industry, for example.

Cars contain many sensors that continually monitor performance and notify the owner when something is wrong, points out Ali Parsa, CEO of digital healthcare provider Babylon. And he says the time will come when we will be able do the same in medicine. “We are approaching an area where we will be looking at our biosensors and signals in our bodies that will tell us what is happening in it. We will know the solution to your problem a millisecond after we have received the signal and that information will be freely available to everyone everywhere.”

Such is the huge promise of digital healthcare: experts say it can revolutionize medical progress, but the pharma and biotech industries cannot deliver these advances alone – they need technology expertise to help meet the growing expectations of their consumers. Industry has already begun to team up with tech firms, both niche companies and giants like Google and Apple, to help it catch up and begin to exploit more fully the field’s almost limitless potential. But at the same time, it is acutely aware of the need to negotiate regulatory and security minefields.

 

Via Giuseppe Fattori

With Ruud Kooi & Annelien Van Boxlaer

AI tools could help us turn information gleaned from genetic sequencing into life-saving therapies. Almost 15 years after scientists first sequenced the human genome, making sense of the enormous amount of data that encodes human life remains a formidable challenge. But it is also precisely the sort of problem that machine learning excels at.

 

Google has now released a tool called DeepVariant that uses the latest AI techniques to build a more accurate picture of a person’s genome from sequencing data. DeepVariant helps turn high-throughput sequencing readouts into a picture of a full genome. It automatically identifies small insertion and deletion mutations and single-base-pair mutations in sequencing data.

 

High-throughput sequencing became widely available in the 2000s and has made genome sequencing more accessible. But the data produced using such systems has offered only a limited, error-prone snapshot of a full genome. It is typically challenging for scientists to distinguish small mutations from random errors generated during the sequencing process, especially in repetitive portions of a genome. These mutations may be directly relevant to diseases such as cancer.

 

A number of tools exist for interpreting these readouts, including GATK, VarDict, and FreeBayes. However, these software programs typically use simpler statistical and machine-learning approaches to identifying mutations by attempting to rule out read errors. “One of the challenges is in difficult parts of the genome, where each of the tools has strengths and weaknesses,” says Brad Chapman, a research scientist at Harvard’s School of Public Health who tested an early version of DeepVariant. “These difficult regions are increasingly important for clinical sequencing, and it’s important to have multiple methods.”

 

DeepVariant was developed by researchers from the Google Brain team, a group that focuses on developing and applying AI techniques, and Verily, another Alphabet subsidiary that is focused on the life sciences. The team collected millions of high-throughput reads and fully sequenced genomes from the Genome in a Bottle (GIAB)  project, a public-private effort to promote genomic sequencing tools and techniques. They fed the data to a deep-learning system and painstakingly tweaked the parameters of the model until it learned to interpret sequenced data with a high level of accuracy.

 

Last year, DeepVariant won first place in the PrecisionFDA Truth Challenge, a contest run by the FDA to promote more accurate genetic sequencing. “The success of DeepVariant is important because it demonstrates that in genomics, deep learning can be used to automatically train systems that perform better than complicated hand-engineered systems,” says Brendan Frey, CEO of Deep Genomics.

 

The release of DeepVariant is the latest sign that machine learning may be poised to boost progress in genomics. Deep Genomics is one of several companies trying to use AI approaches such as deep learning to tease out genetic causes of diseases and to identify potential drug therapies (see “An AI-Driven Genomics Company Is Turning to Drugs”).

 

Deep Genomics aims to develop drugs by using deep learning to find patterns in genomic and medical data. Frey says AI will eventually go well beyond helping to sequence genomic data. “The gap that is currently blocking medicine right now is in our inability to accurately map genetic variants to disease mechanisms and to use that knowledge to rapidly identify life-saving therapies,” he says.

 

Another prominent company in this area is Wuxi Nextcode, which has offices in Shanghai, Reykjavik, and Cambridge, Massachusetts. Wuxi Nextcode has amassed the world’s largest collection of fully sequenced human genomes, and the company is investing heavily in machine-learning methods.

 

DeepVariant will also be available on the Google Cloud Platform. Google and its competitors are furiously adding machine-learning features to their cloud platforms in an effort to lure anyone who might want to tap into the latest AI techniques (see “Ambient AI Is About to Devour the Software Industry”).

 

In general, AI figures to help many aspects of medicine take big leaps forward in the coming years. There are opportunities to mine many different kinds of medical data—from images or medical records, for example— to predict ailments that a human doctor might miss (see “The Machines Are Getting Ready to Play Doctor” and “A New Algorithm for Palliative Care”).

Via Dr. Stefan Gruenwald, Lionel Reichardt / le Pharmageek

This week, MMW was privy to some insights on the future of pharma marketing from Sloan Gaon, CEO of PulsePoint. So what’s ahead in the new year? Here are some of Goan’s predictions: AI and ML will save lives in 2018 AI and ML will revolutionize clinical trial recruitment by finding patients based on a …