Wednesday, October 2, 2019

MyMilk Labs launches Mylee, a small sensor that analyzes breast milk at home

Many expectant mothers are told that breastfeeding will come naturally, but it is often a fraught and confusing experience, especially during the first few weeks after birth. Parents often worry about if their babies are getting enough nutrition or if they are producing enough milk. MyMilk Labs wants to give nursing mothers more information with Mylee, a sensor that scans a few drops of breast milk to get information about its composition and connects to a mobile app. The Israel-based company presented today at Disrupt Battlefield as one of two wild card competitors picked from Startup Alley.

The Mylee launched at Disrupt with a pre-order price of $249 (its regular retail price is $349). Based in Israel, MyMilk Labs was founded in 2014 by Ravid Schecter and Sharon Haramati, who met while working on PhDs in neuroimmunology and neurobiology, respectively, at the Weizmann Institute of Science.

Mylee deviceDuring the company’s stage presentation, Schecter said the device is meant to give mothers and lactation consultants objective information about breast milk.

Breast milk changes in the first days and weeks after birth, progressing from colostrum to mature milk. Mylee scans the electrochemical properties of milk and then correlates that to data points based on MyMilk Labs’ research to calculate where the sample is on the continuum, then tells mothers if their milk is “delayed” or “advanced,” relative to the time that has passed since they gave birth.

The device’s first version is currently in a beta pilot with lactation consultants who have used them to scan milk samples from 500 mothers.

MyMilk Labs already has consumer breast milk testing kits that enable mothers to provide a small sample at home that is then sent to MyMilk Labs’ laboratories for analysis. One is a nutritional panel that gives information about the milk’s levels of vitamins B6, B12 and A, calories and fat percentage, along with dietary recommendations for the mother. Another panel focuses on what is causing breast pain, a frequent complaint for nursing mothers. It tests for bacterial or fungal infections and gives antibiotic suggestions depending on what strains are detected.

Though some doctors believe testing kits are unnecessary for the majority of nursing mothers, there is demand for more knowledge about breastfeeding, as demonstrated by the line-up of breast milk testing kits from MyMilk Labs and competitors like Lactation Labs, Everly Well and Happy Vitals. Haramati said on stage that MyMilk Labs plans to eventually transfer some of the tests’ capabilities to the Mylee.

Molecule.one uses machine learning to make synthesizing new drugs a snap

Say you’re a pharmaceutical company. You’ve figured out that a novel molecule could be effective in treating an illness — but that molecule only exists in a simulation. How do you actually make it, and enough of it, to test in the real world? Molecule.one is a computational chemistry platform that helps bring theoretical substances to life, and it is debuting its product onstage at Disrupt SF Startup Battlefield.

Computational chemistry is, believe it or not, something of a hot ticket right now. The explosion in computing resources over the last decade has made it possible for the extremely complex systems of molecular biology to be simulated in high enough fidelity to produce new drugs and other important substances.

For example, say a company knows that a condition is caused by overproduction of a given protein. By simulating that protein in the soup of the cell environment, computational chemists can also introduce and virtually observe the behavior of thousands or millions of molecules that don’t occur naturally but might, say, lock down those excess proteins and tag them for removal by the cell.

This process of drug discovery has been productive, but unlike in the real world, in a simulation you don’t actually have to make that magical molecule. It’s just a bunch of numbers interacting with other numbers. How can a pharmaceutical company, which may have paid a lot of money for those numbers, turn them into actual molecules? That’s where Molecule.one steps in.

1.4.0 Molecule Dashboard Reaction tree

Essentially, the company has created a software platform that automates the process of getting from chemicals A, B, and C to chemical Z, with the many steps in between accounted for and documented. It’s based on a machine learning system that has ingested millions of patents and known chemical processes, allowing it to connect the dots and propose a method for creating pretty much any complex organic molecule. In other words, once a drug company has the “what” — a molecule or compound that may fight Alzheimer’s — Molecule.one provides the “how.”

Piotr Byrski met co-founder Paweł Włodarczyk-Pruszyński (who goes by Maxus to avoid confusion with COO Paweł Łaskarzewski) while in college, where they studied and did research together, eventually both earning MDs. They discovered a shared aversion to the grunt work of chemistry — beakers, distillates, titration, and so on.

molecule one header

 

“We found out we shared a similar analytical approach to chemistry. A lot of chemists really like the cooking process involved with organic synthesis,” Byrski told me. “I have to say… I never liked it very much. That made me think that there are many things in the everyday life of a chemist that can be automated, and need to be automated.”

“Automating organic synthesis seems like just another difficult automation problem, but it’s one with real effects. Real people are suffering because drugs are coming to the market,” he said. “We thought we could help. So we did some research, and we found that the field is so under-developed —  the direction research is going is completely unsatisfactory. We began market research — we were both first timers so it was pretty new to us at the time — and we found out there was a big market need for this. It wasn’t a scientific discovery that would sit on a shelf, it could be applied today to help multiple industries.”

By the time they were working on this, companies were already applying simulation, and statistical techniques (machine learning is essentially weapon-grade statistical analysis) were already popping up. BenevolentAI started in 2013, Recursion in 2014, Atomwise in 2015; clearly the field was growing, and is still adding new companies, like ReviveMed. But these are mainly focused on the question of new drugs based on simulations.

“They provide a list of maybe tens of thousands of structures to a pharmaceutical company, but the company then needs to actually verify whether the predictions have any real-life backing. For that you need physical access to these molecules — just knowing the structure doesn’t cut it,” said Byrski.

Molecule.one’s system tells them how to manifest these structures.

1.1.1 Molecule Dashboard Compounds

“We are making the whole synthesis pathway, so going from compounds that are available to ones that you want,” said Włodarczyk-Pruszyński. “Along the way we need to solve many problems — there are many reasons why a reaction could fail. We want to tell users how to make compounds with the process with the highest chance of success.”

And succeed they have: “Our system works for structures that have never been seen before by any chemist,” Byrski said.

The obvious question is why these huge pharma companies, with their bottomless pockets and technical expertise, don’t put together their own synthesis platforms. It comes down to people.

“The most important factor is that it’s hard for a pharmaceutical company to hire machine learning specialists who have a deep background in chemistry. Over 90 percent of the people I know that work on this in the pharmaceutical business are chemists who have some training in machine learning. This is a difficult problem that requires coming at it from the opposite direction,” Byrski explained. “Our head of machine learning [Stanislaw Jastrzębski] is a PhD from the computational side, who would normally go to Google, Facebook or Microsoft. We’ve built a team that is unique in how it bridges the computational technology and chemistry.”

The databases used by Molecule.one’s systems, surprisingly, are mostly public. The U.S. Patent office has tons of patents involving chemical processes — some important, some small, some obscure, some obvious, but all verified and presented formally. This was a gold mine sitting in plain sight, Byrski said. Or perhaps a box full of Lego pieces just waiting to be assembled into the right machine.

The main “proprietary” information they used was a private listing of commercially available chemicals and their prices. A molecule may have more than one pathway to reach it, after all, or perhaps thousands, and one might be cheaper than the others or involve fewer toxic reagents.

With strong results from public databases, they have a better chance of getting pharma companies to share their internal databases when signing up for the service.

The actual business is conducted SaaS-wise, naturally, and all the work takes place in the cloud. There’s also an enterprise tier that allows for on-premises operation, for companies that would rather not have their trade secrets anywhere but on company-owned infrastructure.

So far the company has bootstrapped, and currently has about $400K in the bank, which Byrski said should last them well into next year. “The biggest cost is people,” he said. “Developers, designers, chemists. We’re a software business so we don’t have a lot of other costs — we don’t have to hire a lab, for example.” So they are looking for funding to help hire and scale.

It’s not common in Poland to segue directly from medical school into a startup, Byrski admitted. But he and Włodarczyk-Pruszyński felt that this was too significant an opportunity to do good to pass up. With luck their platform will prove as popular as the drug discovery startups that helped make it necessary to invent.

Bithumb to talk to govt for regulated exchange

Bithumb is looking to expand in India by partnering with local cryptocurrency exchanges, fund Indian cryptocurrency startups, and introduce new initiatives for Indian traders https://ift.tt/2pxfWjX https://ift.tt/eA8V8J

ShareChat transparency report says it provided user info to govt on 33 out of 41 requests

Sharechat took down 11.9 lakh pieces of content and banned 5,696 accounts permanently for violating community and content standards and terms of services https://ift.tt/2pnzjvt https://ift.tt/eA8V8J

Political parties go big on Facebook ads as state elections near

The ruling Bharatiya Janata Party (BJP) and its affiliates lead, with collective spending of Rs 21.6 lakh in the three states. https://ift.tt/2pvdVVi https://ift.tt/eA8V8J

Leo Aerospace provides bespoke rocket launches — from a hot air balloon

The demand for orbital launches is increasing steadily, and the industry is nowhere near keeping up. Leo Aerospace thinks it can help with a launch technique that’s more efficient and requires far less infrastructure than an ordinary rocket: a hot air balloon. With a rocket attached to it, of course. It sounds wacky at first, but it could prove to be an economical and flexible way of getting to orbit.

Leo is originally out of Purdue, one of two such teams on stage this week at Disrupt SF Startup Battlefield. Co-founder and CEO Dane Rudy said they were looking into new and better ways to achieve orbit besides the traditional surface-based rocket approach.

leo j“We found this really elegant solution that was actually tested in a rudimentary way in the 50s by the Air Force, which is launching rockets from an aerostat — a balloon,” he said.

Perhaps used to countering narrowed eyes and barely disguised incredulity at this point, he hastened to follow up.

“It actually worked really well for what it was designed for. The issue they ran into was that the U.S. shifted toward sending people to the Moon — so there just wasn’t a need for that technology in the Apollo program. But the rise in small satellites has created a huge demand tailored to these capabilities,” he said.

It turns out using a balloon has big benefits. A large amount of a rocket’s fuel and engineering is dedicated to pushing it from the ground, where the atmosphere is heaviest, to the thinner upper atmosphere, where drag and other issues are much less of a worry. By going the first few miles straight up in a balloon, much less rocket is needed to get into orbit, since you’re skipping one of the hardest parts.

The technique is more or less exactly what you’d imagine: A large balloon inflates and lifts the payload, a small rocket, to a designated altitude. Once there it aligns itself and… well, lifts off is perhaps the wrong term. But it ignites and exits the atmosphere at a planned trajectory and inserts the payload into orbit.

There are already air-launch systems out there that use planes rather than balloons, presenting their own challenges and advantages. Leo Aerospace’s main draws are flexibility and cost.

“Our system is fully mobile — it doesn’t require any ground infrastructure,” said Rudy. “The whole thing fits into a regular shipping container.”

That means it can take off wherever and, perhaps more importantly, whenever the client chooses.

 

shipping

Right now the launch industry is expanding like crazy, both because of an increase in total launches and the rise of “ride-sharing,” where dozens of payloads share the cost of a single rocket. The cost goes down, but there are serious inconveniences.

“They don’t have much choice in when they launch or what orbit they’re going to. There’s also the complexity of having to ride with a bunch of other payloads on board — you have to compromise on timing and so on,” Rudy said.

While ride-sharing means many payloads will get to space that might not have a few years ago, it also means they might wait for years while the rest of the seats fill up and get ready to roll. With Leo it’s practically Domino’s for orbit.

That’s all great in theory, but the fact is no one has made a balloon-based commercial launch system. When the Air Force did it, it was pretty crude: The rocket was carried in a vertical position and shot right through the balloon when it went up. That kind of rules out reusing the balloon, but Leo’s entire business is founded on reusability, since that brings costs down immensely.

launch regulus

“That was one of the big problems we had to solve — the expense of the balloon itself; helium is expensive, and the envelope [i.e. the balloon material] is expensive and fragile,” said Rudy. “How do we make that zero stage, as we call it, reusable?”

Amazingly, they determined that tough, ripstop nylon and hot air were actually the best solution. It’s remarkably close in principle to an ordinary recreational hot-air balloon, but with the slight difference that it has to fly up to 18 kilometers of altitude and carry a rocket with it.

“The difference is how do you control and command this sort of vehicle, integrate it into airspace, suspend the rocket beneath it and all that,” Rudy said. “All the stuff you have at Vandenberg Air Base for a launch — we have to make all that mobile.”

Mike Mojave Launch 11 of 25

A bit like going from an ordinary car to a self-driving one, Leo’s balloon may be similar to the recreational type in its basic form, but the technical advances are in how it is controlled and tracked. They can adjust for wind, control the yaw and rotation, rise to a very precise altitude, and so on — naturally, all remotely and with partial autonomy.

The rocket doesn’t shoot through its balloon as before, but fires off at a mission-determined angle. 18 km closer to space, with far less air resistance to worry about, the three-stage rocket (two solid, then one non-cryo liquid) can be much smaller and have far less mass — requiring less than half the fuel to lift a given mass to orbit. To be specific, the system is specced to send 33 gross kilograms, 25 kg of payload, to a 550 kilometer orbit — or about twice that to 300 kilometers.

December saw the company performing reduced-scale tests at altitude, an important stepping stone to regulatory approval. The plan is to make their first full-scale suborbital launch next year with their first customer’s payload on board. Orbital launches are planned for 2022.

 

Leo has gotten through December’s tests on a quite barebones budget for a space startup of about $520,000, through TechStars and a grant from the National Science Foundation. That’s great for a foundation, Rudy said, but full-scale tests and an eventual transition to commercial operations will take more than six figures.

An Air Force Small Business Innovation Research grant has opened the door to other government sources, and there’s been interest from that quarter in the non-orbital potential of the system, for instance high-altitude testing, mobile communications infrastructure, and so on. So already there are multiple eggs in multiple baskets — an attractive quality for investors.

“We’ve done all the foundational work,” said Rudy.  “Now it’s just about scaling up.”

OzoneAI wants companies to pay you for your data, upending the ad model

Imagine this. Instead of giving away your personal data so web giants can show you ads, you cut out the middle person and allow advertisers to pay you directly for your data.

It’s a novel idea for a new startup that bills itself as a “data privacy” company.

OzoneAI says it preserves users’ privacy by allowing them more granular controls over who gets their data. In the startup’s utopian vision, companies can skip over the major advertising giants like Google and Facebook and buy access to anonymized data from the users themselves. That could mean companies buying your Spotify playlists, your Amazon wish list, or your access to your social media. The user is paid for the access, and the company gets to use the data for better targeting their ads.

The company made its public debut at Disrupt SF on the Startup Battlefield stage.

The startup seeks to capitalize on the wave in recent years of mistrust over online ads. As websites have become more aggressive in their advertising by persistently tracking users and collecting personal data, users have fought back with ad-blockers and privacy apps.

OzoneAI thinks there can be a happy medium between selling and protecting your data. With an app, users can grant companies access to their data and get paid for it. In return, companies get anonymized data which they can use for better targeting consumers.

By taking big tech out of the equation, the startup thinks it can make advertising more efficient for everyone.

But not everyone will see it that way. The privacy-minded with their ad blockers aren’t likely to lower their own bar to allow advertisers any more access to their data than is necessary. But for those who want to make a quick buck, will their actions be motivated more by making money and less about privacy?

When asked, the company’s founders Lyndon Oh and Ben Colman, who both previously worked for Google, denied it was encouraging users.

“However much activity you want to sell, that’s completely up to you,” the founders said. “You are anonymized as an identity; however, your data is not. The fact that you watched ‘The Avengers’ five times last week is not anonymized — you are selling that directly.”

Users’ data is fed into the machine learning engine to make recommendations for other companies and services, allowing the process to match activity to another business. OzoneAI makes its money by charging a 30% cut from the businesses who subscribe to a user’s data, which they say still provides better value than the “spray and pay” model that companies who want to advertise on Facebook and Google are subjected to.

The user can use the OzoneAI app to log in to their various social networks, shopping sites, email provider and anything else. That authorizes OzoneAI to vacuum up your data and begin processing it.

Unsurprisingly, many will turn their nose up at the concept of having a startup they’ve never heard of be effectively given unfettered access to all their online accounts so they can sell your data directly to companies hungry to advertise to you. The startup says it, like any company with similar access to the firehose of a person’s online life, has access to that data but immediately writes the data to its servers in an encrypted way. The user retains the private key to their cloud-stored encrypted data, so OzoneAI says it can’t access the user’s data. Through the app, the user can also “detonate” and destroy any of their collected data.

But by giving users control of their data, the user has to trust OzoneAI more than the existing advertising system. Given how broken the existing system is, it likely wouldn’t take much trust at all. But it’s an uphill battle for an emerging startup with not much of a name. And one hiccup could send that trust tanking.

The founders said other tech companies are “doing the same thing,” and that’s reason enough why users should trust OzoneAI.

But “another company does it” doesn’t mean it’s the right thing. OzoneAI was “just putting users directly in front of the business,” the founders said.

The old saying goes, “If you’re not paying, you are the product.” And yet as much as you are paying — with your data — then you inexplicably are still the product.

Two enterprise-worthy password managers: LastPass and RoboForm

Everyone in your company needs a password manager -- and there are lots of great options. But two cross-platform tools rise above the rest, thanks to their excellent support for enterprise networks.

Twitter and TweetDeck are experiencing partial outages

It’s not just you, Twitter has gone wobbly again. Users of the social network in Asia and Europe are reporting a range of problems tweeting and viewing certain types of content this morning.

Among the problems being reported are not being able to post certain types of content to the site (such as polls and media), though at least some users are still able to post text tweets saying they’re having problems.

Other users aren’t seeing latest replies to their tweets. In my case I’m unable to view latest replies on Twitter’s desktop product but can see them on an (older) version of Twitter’s iOS app.

Some Twitter users are also reporting problems posting to Android.

A Twitter spokeswoman confirmed to TechCrunch it’s having problems — pointing us to a tweet from @TwitterSupport where the company says it’s experiencing outages across both Twitter and its alternative client, TweekDeck.

The problem is also affecting being able to view DMs, per the tweet.

“We’re currently working on a fix, and should be back to normal soon,” Twitter adds, without providing detail about the cause of the issues.

The flakey service comes a few months after a major outage for Twitter.

Back in July Twitter’s service went down for a full hour. In that case an “internal configuration change” caused the issue — which Twitter subsequently rolled back.

It also suffered problems with direct messages in the same month.

Coincidentally or not, the company rolled out a major redesign of its desktop product this summer.

Twitter’s new ‘Facebook-style’ look has not been universally popular, to put it politely. Whether the redesign is the root cause of the recent bout of service flakiness remains to be seen.

Twitter’s status page sheds zero light on the matter — currently reporting that “all systems are operational” when that’s patently not the case.

We’ll update this report with any further details on the problems from Twitter.

Impala builds a single API for the entire hotel industry

Meet Impala, a London-based startup that wants to make it easier to interact with hotel data. The startup is building a layer on top of legacy hotel systems to standardize everything with a modern REST API.

And Impala has just raised an $11 million Series A funding round from Stride.VC, Xavier Niel/Kima Ventures, Jerry Murdock, the partners of DST Global and existing investors. The company had previously raised a $1.75 million seed round.

Essentially, Impala wants to be as simple as Stripe, Twilio or Plaid. With a few lines of code, any developer should be able to get started with Impala before diving deeper.

If you’re not familiar with the tech stack of the hotel industry, hotels use Property Management Systems to manage rooms, room types, pricing, extras, taxes, etc.

“One of the reasons it's necessary is that hotels never replace that underlying system (ever) and so there's no incentive for those old systems to build open APIs (even if they could),” co-founder and CEO Ben Stephenson told me.

Developers working on products in the hotel industry currently have to build a ton of integrations to connect to all the different hotel systems. Impala wants to do the same work once and for all, and standardize the API for anyone building services on top of hotel systems.

In other words, if you want to know how many standard rooms are left in different hotels, you can query those hotels using the same API call. It becomes much easier to manage one or multiple hotels, and build apps, websites and internal services that interact with a hotel system.

With today’s funding round, the company wants to build more integrations with hotel systems. It currently supports 8 different systems, but universal support will be key when it comes to making Impala the universal language of the hotel industry.

Impala is also working on a direct booking API. Right now, many hotels manually upload booking data to Booking Holdings websites (Booking.com, Priceline, Agoda, Kayak…) and Expedia Group websites (Expedia, Hotels.com, HomeAway, Trivago…), or use a channel manager.

Those channel managers act as middlepersons that send information to multiple websites at once. “The problem with this is that if you and I wanted to start a new online seller tomorrow, we would have to connect to all of the different channel managers,” Stephenson said.

A direct booking API would lower the barrier to entry for Expedia and Booking competitors. It would also open up possibilities for new types of players who don’t necessarily sell hotel rooms today. You could imagine being able to book a room directly from a city guide website, a conference website or a music festival app.

It wouldn’t be a Booking.com embed, it would leverage Impala’s direct booking API to book directly with the hotel, which would lead to reduced commissions.

Now, gamers can get university degrees in esports

On their first week in class, a group of students is playing a first-person shooter video game in a sleek new digital studio.It's their introduction to the degree in esports they've all enrolled in. https://ift.tt/2mVrP28

Ex-Yahoo worker pleads guilty on hacking account for nudes

A former Yahoo software engineer has pleaded guilty to hacking personal accounts of thousands of Yahoo users searching for their sexual images and videos. https://ift.tt/2opsZ6e

Visa, Mastercard reconsider backing Facebook's Libra: Report

Visa, Mastercard Inc and other key financial partners may reconsider their involvement in Facebook Inc's cryptocurrency, Libra, the Wall Street Journal reported on Tuesday, citing people familiar with the matter. https://ift.tt/2onYJsA

Trump administration cannot bar states from passing own internet rules, says US court

A US appeals court rejected the decision of the Federal Communications Commission to declare that states cannot pass their own net neutrality laws and ordered the agency to review some key aspects of its 2017 repeal of rules set by the Obama administration. https://ift.tt/2pfUCiF

Tuesday, October 1, 2019

Super Mario Maker 2 catches up with 1998, lets you play online with friends

Though Super Mario Maker 2 was the first game in the decades-old Nintendo series to support online multiplayer, the feature premiered in late June with a curious twist. Its owners could jump online and play a slew of custom-created levels in either versus or co-op modes, but only against strangers.

Shortly before the game's Switch launch, Nintendo acknowledged how crazy this sounded and promised that friends would eventually be able to pair up in these online modes via friend lists. In the months that followed, Nintendo remained utterly silent... until the wee hours of Tuesday night, when the game's 1.1.0 patch went live.

We can confirm that Super Mario Maker 2 now works like most every other online game we've ever played. We were able to contact people we knew on a Nintendo Switch friend list, start a SMM2 session, and play with (or against) said friends. (Only Nintendo could merit an entire article about playing an online game with friends in 2019.)

Read 5 remaining paragraphs | Comments

https://arstechnica.com

Arizona's Maricopa County is set to have the second largest concentration of US data centers by 2028, as the state races to increase electricity production (Pranshu Verma/Washington Post)

Pranshu Verma / Washington Post : Arizona's Maricopa County is set to have the second largest concentration of US data centers by 202...