The Freebird Solution

Freebird has two separate offerings – for leisure travellers (self-managed) and for businesses (managed). After having bought their flight ticket, self-managed travellers can go to the Freebird website and ‘purchase’ the Freebird service. The travellers enter the details of their flight as well as their personal details provided while booking the ticket. This information helps Freebird track the flight for disruptions and keep the traveller’s details ready in the event of a rebooking.

For the business offering, Freebird works directly with corporates and Travel Management Companies (TMCs). While booking a ticket, travel managers can ‘apply’ Freebird to an employee’s travel. In the event of a disruption, the employee is able to rebook the flight using Freebird.

Freebird’s mobile flight rebooking solution allows travellers facing disruption to rebook tickets with 3 simple taps without an app or download. When a flight is disrupted, Freebird sends the traveller an alert via email or text with a link. Upon clicking the link, a browser tab opens up with a list of all available flights to the traveller’s destination ordered by time of arrival. The traveller can select the option that works best. After selecting the desired flight, the next screen shows the traveller his/her booking details that were provided while purchasing/applying Freebird. After reviewing it, the traveller can book the flight by pressing ‘book flight’ at the bottom. Once the ticket is booked, a confirmation email with the ticket is sent to the traveller. With 3 simple taps, the traveller can successfully rebook a flight without the hassle of waiting in line or trying to reach someone at a call center.

For their service, Freebird charges a prior fee. Currently, they are charging a fixed promotional fee of $19 for a one-way flight and $34 for a roundtrip with a dynamic pricing model in the works. In the event of a disruption, which the company defines as a cancellation or delay of over 4 hours or a missed connection due to a delay, Freebird pays for the traveller’s new ticket. If there is no disruption, Freebird keeps the money. As flights run mostly without issue, the company gets to make a profit, apart from averaging out the pay-out for new tickets. The dynamic pricing model will help price the fees more appropriately after considering the risk of the flight based on numerous factors such as origin, destination, duration, and seasonal factors.

Not a travel insurance

While Freebird’s model may make it seem like travel insurance, it technically isn’t and is different in many ways. Travel insurance covers a wider range of events such as medical, hotels and baggage loss among other things, while Freebird only focuses on getting the traveller a new ticket. As no additional payments are required once Freebird has been purchased, travellers avoid the temporary burden of expenditure and the hassle of filing claims and getting reimbursed that come with insurance. On the pricing front, domestic travel insurance costs upwards of $30 while Freebird is priced significantly lower at $19.

Freebird has been mindful of the dynamic nature of travel. Apart from accommodating changes in the itinerary, Freebird also allows travellers to rebook to alternate airports arriving at the same city. To cover instances where a flight is disrupted and Freebird hasn’t sent the traveller an alert, a dedicated support team based in the US has been put in place to respond to travellers. Travellers can also get frequent flier miles on their rebooked tickets by providing Freebird with their frequent flier number. However, Freebird’s services are currently available only on domestic flights and have to be purchased at least two days prior to departure.

Flying Into The Future

A 2010 study by Berkeley estimated that flight delays and cancellations cost the US economy over $32.9B (2007), $8B shy of a Senate Joint Economic Committee’s estimate of $40.7B (2007). While airlines and passengers are the two parties directly involved, businesses face an adverse impact in terms of increased costs and productivity losses.

Freebird is focusing on the business travel market to drive growth. Over the past 18 months, the company has tied up with over 100 corporate clients and 10 TMCs such as BCD Travel, Adelman, Atlas Travel and Altour among others. Freebird’s API developed in 2016 helps integrate their services into the partners’ existing infrastructure without causing any disruption.

In October, Freebird raised $8M in a series A round led by American Express Ventures. The round saw participation from new investors such as CitiVentures and PAR Capital Ventures as well as existing investors General Catalyst and Accomplice. The team intends to deploy these funds to grow their business travel offering. At its current pace, the company is on pace to protect over 250,000 travellers annually. Freebird’s approach to solving the disruption problem has won itself traction and recognition in the industry from the likes of Business Travel New and Skift.

The post Freebird – Disrupting Flight Disruptions appeared first on Techweek.

The Cambridge-based startup’s share rates are high as it’s invested in promising research that seeks to find personalized vaccines and therapeutics for 21+ diseases using messenger-RNA (a nucleic acid). Yet, it’s important to note that only 10 have entered clinical studies and none of the potential cures has received much traction in the FDA pipeline.

“We’ve continued to show our investors progress, as well as setbacks, but everybody has setbacks in research,” said Moderna’s Chief Financial Officer Lorence Kim. He believes that such a trajectory has “kept investors happy.”

Getting IPO-ready

Moderna kicked off its IPO debut at a time when it is said to be saddled with $360M+ in operating costs amassed in the first nine months of 2018.

Yet, the startup is currently already quite cash-flushed. As of September 2018, with $1.2Bn in cash (and no debts). It also amassed $2.6B in total funding from its strategic collaborators and investors. In fact, its total revenue was $100M+ in the first nine months of 2018 or more than a quarter of total expenses.

But as the capital-burning research trials progress, they will definitely need more money to function in the future. Also owing to the wavering investor sentiment for the sector, what with the recent S&P Biotechnology Select Sector Index selling off 20%+, it would be intuitive for Moderna to keep its options open.

This year, a number of other biotech companies have decided to go public including Singapore-based Aslan Pharmaceuticals (raised $42M), China-based Innovent Biologics (raised $421M), California-based Equillium (hopes to raise $86M).

It is believed that biotech startups need $2Bn+ and 12+ years just to develop a drug, and becoming profitable tends to be longer than the life cycle of a typical biotech venture fund. By going public, VCs can hope for earlier returns and the startup can add to its reserves for its drug-making journey.

The intent to go public, coming 8+ years since Moderna’s inception, makes it one of the rare biotech startups to stay private for so long. According to Moderna Therapeutics’ CEO Stephane Bancel, this has helped them take risks and try out new things (such as working on multiple projects at the same time) without worrying too much about public scrutiny or pressure.

“We very consciously decided to stay private to be able to turn on a dime based on what we learn in the labs initially,” Bancel told Xconomy. “That has served us well, because over the 6 years the company has been operating, there have been a couple of bumps on the road on the science—and nobody panicked.”

The Road to the Future of mRNA

This proverbial road to achieving greatness in science, particularly in mRNA therapy, first started in 2010. It was in this year that Harvard University scientist Derrick Rossi manipulated mRNA to deliver proteins that turned adult cells into embryo-like stem cells. Seeing potential business value in this solution, Harvard cardiovascular scientist Kenneth Chien and serial entrepreneur Robert Langer joined forces with Rossi to pitch a new stem cell company to Flagship Pioneering, a venture capital firm.

At that time, Flagship’s CEO Noubar Afeyan felt that the proposers need to try something different, He suggested that they seek out ways to use mRNA as a protein therapy vehicle. Subsequent experiments, on mice, showed that modified mRNA could produce proteins in the liver that would then be circulated through the bloodstream. With this academic proof, Afeyan became an instant believer and joined the trio to setup Moderna in the same year.

Now, to understand exactly what Moderna does, here is a quick biology lesson: mRNA translates the genes of DNA into dynamic proteins that can be used to treat diseases.

“Because mRNA is nothing more than a copy of DNA, you could potentially make any protein you want, because the code is there,” said Afeyan in an interview with CNBC.

Typically, biotech companies spend a lot of time and money cooking up these proteins (as drugs) using genetically engineered cells inside large industrial vats. Moderna turns the industry on its head, by injecting the diseased with individualized synthetic mRNA that turns the body into ‘little drug factories’. So, when this injectable mRNA enters the body, it delivers instructions to diseased cells to generate specific proteins that can, in theory, treat the disease. And even though it’s called a vaccine, mRNA therapies are not built to prevent illness but to coax the immune system to recognize and attack certain markers on affected cells (such as cancerous cells).

What’s more, this kind of a solution can be used just to fight more diseases than just cancer. In fact, Moderna already has (early stage) clinical trials ongoing for diseases such as CMV, HMPV/PIV3, Influenza H10 and H7, Zika and Chikungunya.

Even with so much going in, for the first two years, Moderna kept all their work in the dark, without any talk of their trials or successes. This move faced a lot of criticism, as they were seen to be withholding information. But they soldiered on undeterred. Moderna spent the time filing broad and deep intellectual property. This ensured that even if other companies succeed in using mRNA-based techniques in areas not yet considered by Moderna, it could still enjoy royalties.

Out of Stealth Mode

In 2012, Moderna announced that they had raised $40M in funding and started publishing papers about the technology it was developing. This was followed by a spree of successful fundraising efforts. In March 2013, Moderna inked its first big pharma partnership with a $240M investment by AstraZeneca to co-develop mRNA therapies that can stop heart attacks along with metabolic, and renal diseases. October that year, it raised a grant worth $24.6M from the Defense Advanced Research Projects Agency (DARPA) for creating therapies to combat infectious diseases. Then, in November it also raised $110M in new equity financing from undisclosed sources.

To ensure that it doesn’t spread itself too thin working on various drugs and its platform, in 2014 it created a venture unit to start a bunch of subsidiaries – Valera, Elpidera, Caperna, and Onkaido. Each of these had different management teams and employees to manage different sets of its mRNA drugs. While this move helped the brand make progress on several therapies at the same time, with risks hedged, it was too complicated for Moderna and its investors to wrap their minds around. So in September 2017, they shut this down.

Meanwhile, in January 2015, the then-preclinical biotech company secured $450M in financing. And then in July 2016, a partnership worth up to $315M to Moderna was announced, with Boston-based Vertex Pharmaceuticals Inc. to combat cystic fibrosis.

“We’re playing a very long game,” Bancel said. “Our goal is to bring the best medicines to patients, and we always like to start with a partner who can bring expertise to a new area we’re looking at. It’s about whatever organ system we can get messenger RNA into.”

It also began clinical trials for solid tumors, in 2017, with pharma giant Merck (signed partnership in 2016 for $200M). And by November 2017, a cancer vaccine was injected into its first patient.  

Most recently, in its S-1 report, Moderna states that it raised $560M in a Series G funding round in January and February 2018, followed by $125 million in a Series H funding round in May 2018. What’s interesting to note, is that even with no marketable product (vaccine) in hand, it’s still garnering quite a bit of investor attention.

It has also invested some of these funds into setting up and hiring for its new GMP clinical development manufacturing plant, in Norwood, Mass, that opened in July 2018. The 200,000-sq. ft. facility, has been designed to produce materials for pre-clinical, Phase 1 and Phase 2 programs. It also features state-of-the-art facilities including an army of machines and robotic handlers.

According to a report in the American Chemical Society, in the last five years, the 645 employee-strong Moderna has spent $450M on research. And it plans to spend another $500M across the next five years.

Their research is expected to eventually solve the startup’s issues with creating synthetic mRNA drugs that are safe for human consumption, while also being able to generate the right quantities of protein to be effective against the disease.

Towards a Bigger and Better Moderna

Getting the research right would be the first step towards Moderna’s aim to become a company worth more than $50Bn – something very few biotech companies (such as Amgen, Gilead Sciences, and Biogen) have achieved.

But there is enough room and more in the market to make this goal possible, as the global mRNA vaccines & therapeutics industry is expected to stand at more than $115M by 2020.

Others such as CureVac, Translate Bio, and BioNTech are as well are working on similar mRNA-based solutions. But truly unique about Moderna is the fact that for the longest time while remaining privately held, it has secured $1.6Bn+ in venture financings and $1Bn+ through partnerships. So the public sentiment is strong with this one.

Though it’s still to see the quantifiable success of its therapies, Moderna remains confident of its progress. “We built and continue to invest in a platform to advance the technological frontier of mRNA medicines,” Moderna said in its S1 filing report. “We made and continue to make forward investments in scalable infrastructure and capabilities to pursue a pipeline of potential medicines that reflect the breadth of the mRNA opportunity.

The post Moderna Therapeutics – Largest Biotech Company to Go Public appeared first on Techweek.

The problem

As Shepherd Therapeutic’s own website points out, cancer isn’t one disease, but many. There are 300+ discovered forms of cancer, and each cancer requires its own very specific set of clinical trials, and its own specific therapeutic or cure.

Now rare cancers are cancers that afflict, or are diagnosed in, fewer than 6 in 100,000 people each year, and they include sarcomas, and brain, oesophageal, cervical, prostate and most paediatric cancers, among others. A 2010 study published in the journal Public Health Reports found that “rare cancers” accounted for 25% of all cancer diagnoses in the US, and a 2017 study in the Indian Journal of Medical Research says that rare cancers account for 22% of all cancer diagnoses worldwide.

It becomes particularly difficult, in the current medical and economical climate, to develop cures for rare cancers for a variety of reasons. Most of the funding that goes into cancer research is poured into the more commonly detected types of cancer, like breast cancer, lung cancer and skin cancer, while rare cancers get pushed to the sidelines. Pharmaceutical companies, which drive a lot of the research around cancer, are also less motivated to work on developing cures for rare cancers, because the market for such cancers is much smaller than other, more commonly diagnosed cancers that affect a numerically higher number of people each year.

There are also the inherent difficulties in carrying out trials and collating existing and new information about rare cancers given the very small patient population to conduct trials on.

Given the lack of research, understanding and available cures for rare cancer, in 2006, a group of Italian researchers actually found that the survival rate for rare cancer patients was nearly 18% less than for patients with common cancers.

Shepherd Therapeutics claims that there are 100+ rare cancers for which there just exist no cures at all. They believe that millions of people affected by rare cancers who could be saved–if only adequate research and development was being done to find therapeutics for their kinds of cancers–shouldn’t simply be left to die.

Shepherd Therapeutics’ solution

Shepherd Therapeutics approach looks to combine “traditional statistical methods and machine learning techniques” with the “decades of combined experience [their] scientists bring” to the current medical and business landscape around cancer research in order to speed up the drug discovery and development process. In a nutshell, it collects and identifies viable research around rare cancers, and then contracts scientists and labs to develop therapies for them.

It basically aims to create the deepest informational insight into rare forms of cancer, and through the collection and creation of these knowledges and insights, to speed up and lower the cost of the process of getting therapeutic projects into labs, hopefully resulting in cures being developed for patients.

The process begins with collecting as much qualitative and quantitative information as possible about each cancer, which the company refers to as “mapping the mutational landscape”. Shepherd Therapeutic claims that it collects data that other drug development companies don’t have the time or energy to collect, including mutational burdens, patient profiles, natural history, optimal cell lines, clinical trial histories and business data around existing research.

Their in-house computational system (DELVE) and its laboratory engine (ADA) stores every piece of data that Shepherd Therapeutics collects, and connects existing information resources using AI and machine learning. It adds qualitative variables to quantitative data, attempting to identify target overlaps between different rare cancers, and comparing experimental data from its labs with existing literature, to potentially eventually lead to the automation of identifying potentially synergistic drug combinations through research.

It also bridges gaps in the knowledge its computational system creates using, wherever possible, literature surveys, discussions with researchers and mathematical deconstructions, using both technology and human ingenuity to accelerate the process of finding cures for rare cancers.

The company and its unique founder

After completing a liberal arts major at St John College in Annapolis, David Hysong was hit by a 12-ton bus whilst investigating child slavery in Cambodia. Whilst recovering from this accident, he applied to Harvard Divinity School (on a dare), from which he graduated in 2009. He was then chosen through a special program to join the Navy SEALS, shortly after which he was diagnosed with a lung infection called SIPE (Swimmer Induced Pulmonary Edema). Further medical attention into this infection revealed that Hysong was suffering from a rare and incurable form of head and neck cancer, called Adenoid Cystic Carcinoma (ACC).

Being told that there is no cure to his form of cancer only inspired Hysong to leap into action, to use everything he had to make a difference to the medical landscape as it stands today, and to make sure, to paraphrase a Dylan Thomas quote that once featured on Shepherd Therapeutics’ website, that rare patients “did not go gentle into that good night”. He leveraged his Harvard network and partnered with former executive vice president of Genzyme, Eugene WIlliams, to form Shepherd Therapeutics in 2015.

Intriguingly, Hysong claims that Shepherd Therapeutics is the only “all female” biotechnology company he knows of, and that his core research team is made up entirely of women. He also claims that his team members are driven largely by passion for the cause, and that some have taken 75% pay cuts to be part of the project.

As of January 2018, Shepherd Therapeutics has raised $6.5M in funding, and has external collaborations with the National Cancer Institute and the Massachusetts Institute of Technology. When asked about Shepherd Therapeutics’ competition, Hysong told Biospace that while there are 200 biotech companies that mention that they are working on one or more forms of rare cancers, there is no other single platform systematically working on developing cures for different types of rare cancers.

As for the future, Hysong says that they aim to expand their research products and ecosystem, and in the long term, to become a “fully integrated pharmaceutical company.”

The post Shepherd Therapeutics wants to ensure cancer patients aren’t left to die appeared first on Techweek.

Computers typically need a long time and hundreds or thousands of examples to understand concepts (Think millions of images and multiple processors for a computer to learn that whiskers are different from a tail).

But these researchers, using a model called the Bayesian Program Learning framework, built an algorithm that gets machines to learn the way humans do: using only a few examples. They used probabilistic programming – code that deals in probabilities rather than specific variables— not only to recognize but also reproduce over 1500 characters written in many languages such as Tibetan, Gujarati, Sanskrit, among others.

At the end of this experiment, the researchers victoriously declared that “fewer than 25 percent judges” could distinguish between the handwritten characters with those drawn by the computer. Euphoria!

But once the chatter around the research died down, it became obvious that if computers can recognize handwriting today, there’s a lot more they will be able to do tomorrow.

While this study was purely academic, other researchers, stirred by the revelation, have been finding ways to make a buck out of what the 18th century statistician Bayes formulated and what probabilistic programming can do.

In February 2017, a startup called Gamalon emerged from stealth mode (it was started in 2013) to reveal some of its own plans based on the Bayesian Program. And last month, this Boston-based startup had raised $20 million in a series A funding led by Intel Capital, with participation from .406 Ventures, Omidyar Technology Ventures, Boston Seed Capital, Felicis Ventures, and Rivas Capital, driving its funding to a total of $32.1 million.

The New Wave of AI

Most data is unstructured. It’s just blobs of text full of what’s called data dirt. Unstructured data cannot be visualised, nor can it be used for machine learning or to run business processes. It can take many computers months to access millions of data points, and even then, the information that churns out may or may not be useful.

But Ben Vigoda, an MIT-trained computer scientist and Gamalon’s CEO says he has a quick-fix.  

CEO Ben Vigoda

In a talk, Vigoda gives the example of a grocery app that connects drivers to grocery stores to pick up supplies for consumers. Say, you need the 12 ounce diet cola but the app, having sent a signal to every cash register’s data system, received results populated with other types of colas. Vigoda says this is a difficult problem to solve with AI because the machine is not educated about the different brands of beverages, the variety among brands, or that ounce can be abbreviated as OZ. This is where Gamalon’s products with its probabilistic programming come into the picture.

Vigoda told Bloomberg last year, “You can run our software on a laptop, and it takes 100 times less horsepower to find an answer.” And companies do. They use Gamalon’s products – Structure, which picks up concepts from raw text efficiently and can adequately describe the product, and Match – which categorizes data and learns concepts quickly. For example, Match will learn the numerous abbreviations, brands, and types and correctly seek out the 12 ounce diet cola.

Vigoda had earlier founded a company called Lyric Semiconductor that did projects for the Department for Defense and used probability to guess user behaviour, purchase patterns, improve results etc. commercially. It was acquired by Analog Devices.

Pictionary with Probabilities

Another idea Gamalon is using probabilistic programming is for its easy-to-create drawing app.

Recognizing drawings through deep learning

In 2016, Google launched its “Quick, Draw” app which used deep-learning to recognize drawings. However, the sketch of what you drew had to be similar to what it had already seen. Gamalon, on the other hand, used Bayesian Program Synthesis to teach the machine concepts such as a line, triangle, rectangle, etc., along with the shapes it could make like a lamp or a chair. The app then used probabilistic programming to recognize key features. As long as the shapes remained the same, the app, Gamalon claims, could guess correctly. It also corrected your drawing of say, a triangle if your hand wavered midway.  

In 2017, MIT Technology Review’s listed Gamalon over companies like Facebook and Tesla as the 21st Smartest Company, crediting its technology’s efficiency advantage over other machine-learning methods. Its efficiency comes in from its probabilistic programming algorithms that learn from less-data, perhaps just a few examples.

But along with groceries or the drawing app, the startup uses its technology to help e-commerce, manufacturing and other companies structure and match text data from disparate sources, such as inventory databases. It can perform numerous roles from optimizing supply chains to doing competitive price analytics, but it’s its role in recognizing speech command that’s most intriguing.

Conversing With Machines

Today’s virtual assistants and chatbots search through large amounts of text and then follow assigned rules in order to respond to questions. But what if they could understand language, make guesses, be less confused, and have conversational memory? Vidoga told MIT that Gamalon has built an interface that defines a tree of conversation and the “various different ways the dialogue might unfold”. This software has far-reaching commercial and practical applications and takes Vigoda closer to his dream of using AI “to get all the participants quickly and seamlessly connected (on a conference call), stay connected, (when the assistant) takes notes, captures hand-drawn figures and re-draws them beautifully, and sends everything to everyone in a nice summary afterwards.”
But AI, like anything with an eye on world domination, is divided into camps: those who are Bayesian like Vigoda, follow probabilistic programming, or those that use neural networks, or reinforcement learning, among others. The future can be none, either or a combination of them all. But for now, Gamalon’s less-data approach, that can also convince privacy campaigners to disincentivize companies from accumulating tons of personal data, to its focus on teaching machines the way humans learn can have significant impact on how we perceive AI and the new ideas that leap in from the periphery.

The post Gamalon’s AI Software Learns Like Humans Do appeared first on Techweek.