size1one posted:

There are many companies working in this space and theranos isn't the only one making big claims. Pathway genomics got called out by CBS last night for exagerating too. It's a large market, that won't go anywhere as along as people exist. The liquid biopsy cancer market alone is estimated to have a cap around 20 billion dollars. 100s of millions (maybe billions at this point?) are being thrown at companies to commercialize lab tests. It's not too surprising that some are trying to get a first mover advantage, even if that means exaggerating their efficacy.

But yeah it was very obvious to other people in the field that Theranos was full of poo poo. There are some physical constraints on these problems that require a full blood draw and you can't just magic them away. There are some paths forward but are likely many years, maybe decades, away from being clinically viable.

size1one posted:

Full disclosure, I work for a competitor.

They were exaggerating quite a bit. Current sequencing technology is too noisy to detect circulating tumor DNA reliably. There just isn't enough ctDNA in the blood at early stages. Either the cancer is further along (at least stage 2) or the test will require more blood & sequencer time. Both of the latter drive up the cost of the test. Based on the prices on their website, they probably are either taking a HUGE loss or don't have a reliable test.

size1one posted:

I wouldn't want to be working for someone making misleading claims that's for sure, but it's a great sector to be in. It's technology that will have big impacts on peoples lives even with it's current limitations. There will be some big winners regardless of a tech bust.

Avalanche posted:

Would agree with you if it wasn't the fact that upgrading an OS in an enterprise environment or even just installing basic rear end updates can sometimes non-stop break some old as poo poo programs the enterprise relies on for day to day usage. Yea, it's more of a problem of companies running programs with code being held together by scotch tape than the OS though.

The issue becomes less severe in non-networked equipment like a lot of portable medical imaging equipment. What is the point in upgrading an endoscope system for example (something used to look down a person's throat) if the company software works flawlessly on XP, the image quality looks good on XP, and the only result in upgrading the system to win 10 is the possibility of incompatibility issues, and performance issues with the systems older hardware?

I don't know if something exists already, but it would be cool if a future version of windows incorporated VMs that ran off of virtual hardware configurations so all a companies dogshit legacy software could be spooled up in separate VMs running older OS versions independent of the higher level OS.

Slanderer posted:

Wasn't this basically Windows XP mode in Windows 7? I mean, you had to download it separately, but once installed it let you install XP programs in the XP VM and launch them from 7 (which handled all of the VM stuff in the background).

NihilismNow posted:

It won't be. Unless you are very lucky your software will be using a a ton of deprecated API's from 1996.

Irradiation posted:

I dont want to defend this guy but this is the correct thing to do. Too many undergrads are just trying to get in a lab so they can toss it on their resume and PIs have wised up to this. Now (good) professors will start asking about what your scientific contributions were to the group you worked for if you're interviewing for a graduate spot.

Lab chores are everyone's responsibility not just some undergrads.

Spazzle posted: