Author Topic: Investment Thesis : computational drug development  (Read 1336 times)

ctuser1

  • Handlebar Stache
  • *****
  • Posts: 1741
Investment Thesis : computational drug development
« on: December 01, 2020, 10:57:24 AM »
I read this today morning:
https://www.nature.com/articles/d41586-020-03348-4

That got my imagination going on one of my pet theses - the biotech revolution. I think it will be an order of magnitude larger than the internet revolution, and will create a lot more value.

While it is difficult to predict or even imagine all the future applications - the fact that DeepMind is able to protein fold so effectively seems to indicate that we are at the cusp of being able to "compute" for the next drug.

I have almost 0 background in genetics or even CS-adjacent fields like bio-informatics. Experts in this field (or adjacent fields) - am I just dreaming up a pie in the sky? or are we really close to practical application of bio-informatics that will soon show up as stock market returns.

------------------

Admittedly, my investment thesis is not well developed.

As of right now, I think the best way to invest in this is to still bet in on the tech sector. I tend to think they will capture a larger part of the value created in the biotech revolution than the old-school pharmaceutical companies (the same way Apple gets most of the profit even though Foxconn manufactures the Iphones).

I am sure new disruptors will emerge. The most conservative way to catch them is likely just index investing on the entire market.

So my crystal-ball investment thesis is a small portion of your money riding on the tech sector (I, personally, just throw it in some individual stocks) + a larger portion of the portfolio riding on the whole market ETF.

-----------------

Comments? Suggestions? Contrary opinions are especially welcome.

bwall

  • Handlebar Stache
  • *****
  • Posts: 1220
Re: Investment Thesis : computational drug development
« Reply #1 on: December 01, 2020, 12:42:15 PM »
Look into the CRISPR-Cas9 technology. I think it's a lot closer to development and implementation.

The two women who discovered this in 2012 won the Nobel Prize this year in Chemistry. When was the last time that someone won the Nobel Prize within 8 years of their discovery? That is a good indication of the potential for this field.

To my knowledge there are now currently three companies that are developing this technology: Editas (EDIT), Intellia (NTLA) and CRSP Therapeutics  (CRSP).

If that doesn't quite strike your fancy, then look into genetic sequencing and writing; Illumina (ILMN) and Twist Bioscience (TWST).

It's worth enrolling in a community college to understand the technology so that you can be a better investor. Most guys on Wall St. don't understand what these companies do, so they hire people with science backgrounds to explain it to them. These scientists may or may not be able to recognize a market opportunity--they're trained in science, after all, not in securities. There is a lot of opportunity here for someone who understands both the science and the stock market.

MustacheAndaHalf

  • Walrus Stache
  • *******
  • Posts: 6652
Re: Investment Thesis : computational drug development
« Reply #2 on: December 01, 2020, 12:53:04 PM »
While watching for Covid-19 news, I often see CRISPR developments like these:
https://www.genengnews.com/news/crispr-technique-effectively-destroys-metastatic-cancer-cells-in-living-animal/
(which reinforces what bwall said, unless we're using the same news source!)

You can sign up for daily news updates, at no cost besides ignoring the industry-related ads.  The ads actually reinforce the idea researchers and hospitals are the intended audience.

Abe

  • Magnum Stache
  • ******
  • Posts: 2647
Re: Investment Thesis : computational drug development
« Reply #3 on: December 02, 2020, 11:59:12 PM »
Not close to this translating into anything other than speculative "investing" in stocks, unless you're planning to retire in 30-40 years. Finding molecules that interfere with your favorite protein is the easy part. If someone told me they found a molecule that inhibits a given protein based on a folding model, I'd like them to show me the following:

1. This is a true effect (likely).
2. This effect occurs in human cells and doesn't result in immediate death. (Unlikely)
3. This translates to the desired effect in human cells. (Very unlikely)
4. This effect also occurs in animal cells. (Likely if above met)
5. This effect causes a desirable effect in the animal. (Unlikely, even if above met).
6. The drug isn't toxic to humans (unlikely).
7. The drug has the desired effect (very unlikely).
8. It is also profitable to make and market. (Unlikely)
9. It does something that the generic drugs don't do. (Unlikely)

Only about 1 in 10,000 drugs that get through step 2 get through the remaining steps. That's where the true cost comes in.

Crispr-cas9 use is also the easy part (see above, but replace drug with modified transcript). My mentor's lab demonstrated that finding in the above press release in animal models (for colorectal cancer & melanoma) about 4 years ago now. That, again, is the "easy" part. Also, I wouldn't trust anyone who says their technique has "no side effects". That may be true in the short time-span they observe the mice (and as far as we understand mice physiology), but again there are so many things that seem to be well tolerated (transiently) in mice and are not in humans. I don't know who runs the PR department for some of the research institutes in Israel, but they're routinely putting out this nonsense and it is frankly embarrassing. The people who are making major contributions in biotech from Israel and elsewhere are publishing papers and keeping very circumspect about what promises are made.

That being said, investing in a whole-market ETF seems wise. Weighing it tech-heavy is probably not, for unrelated reasons related to volatility tolerance and sector risk.

Illumina has been around for 10 years at least, and their sequencers are decent. (Definitely makes my research a lot easier). Again, sequencing is now the easy part, trying to find a use for that information is harder even with very complex analytic analyses. 

I am barred from directly investing in drug companies due to conflicts of interest with my position, but if I could the only ones that would be worth investing in are the large pharma and hospital supply companies. The others are highly volatile, mostly because people lose sight of the fact that healthcare isn't the same as software development or engineering.
« Last Edit: December 03, 2020, 12:11:55 AM by Abe »

Green_Tea

  • 5 O'Clock Shadow
  • *
  • Posts: 89
  • Location: Europe
Re: Investment Thesis : computational drug development
« Reply #4 on: December 03, 2020, 01:33:47 AM »
Very interesting Abe, thank you for the insight!

ctuser1

  • Handlebar Stache
  • *****
  • Posts: 1741
Re: Investment Thesis : computational drug development
« Reply #5 on: December 03, 2020, 08:09:06 AM »
Thank you for the insight.

bwall

  • Handlebar Stache
  • *****
  • Posts: 1220
Re: Investment Thesis : computational drug development
« Reply #6 on: December 03, 2020, 12:31:41 PM »
@ abe; thank you for the insight. Curing cancer is indeed a sticky wicket. Lots of problems need to be solved before we get individualized therapies that target only the cancer cells. One day it might be possible, depending on the scientific breakthroughs that we can make, such as CRISPR-Cas9.

In the meantime CRISPR-Cas9 can be used on diseases other than cancer. Over the summer CRSP released preliminary Phase 1/2 trial results for two genetic diseases:

-Beta thalassemia: Two patients are transfusion independent at 5 and 15 months after CTX001 infusion; data demonstrate clinical proof-of-concept for CTX001 in transfusion-dependent beta thalassemia-

-Sickle cell disease: Patient is free of vaso-occlusive crises at 9 months after CTX001 infusion

You can read more about the trial: https://crisprtx.gcs-web.com/news-releases/news-release-details/crispr-therapeutics-and-vertex-announce-new-clinical-data

They are also targeting Type 1 diabetes and cystic fibrosis among others, basically any disease that has a simple genetic component. Read more about the diseases being targeted here:
http://www.crisprtx.com/assets/uploads/CRISPR-Tx-Corporate-Overview-November-2019.pdf

And, while I'm at it, I guess I should disclose that I'm long CRSP stock. Also, the stock is up today 10% (or so) on no news, which makes no sense to me.

J Boogie

  • Handlebar Stache
  • *****
  • Posts: 1531
Re: Investment Thesis : computational drug development
« Reply #7 on: December 03, 2020, 01:12:00 PM »
Anyone familiar with Zymeworks? I have traded in and out but now just have a few shares. I will probably buy dips moving forward but my thesis is that their core value is their molecular modeling software for optimizing protein structure.

I figure if they don't achieve decent success on their own that will make them a solid acquisition target at least, given the proof of concept they've shown with the clinical trials of their ZW25 monotherapy.

ChpBstrd

  • Walrus Stache
  • *******
  • Posts: 6712
  • Location: A poor and backward Southern state known as minimum wage country
Re: Investment Thesis : computational drug development
« Reply #8 on: December 11, 2020, 11:46:51 AM »
I think the idea is that using computers to test billions of chemical combinations at a time in a virtual cell environment will change the 1 in 10,000 math for today's manual processes so that maybe the computers narrow it down to a top-100 list, or a top-10.

Then, a separate program could calculate several optimal pathways to manufacture the compound.

This would be revolutionary, particularly since the patent environment is set up to subsidize companies for taking 1 in 10,000 risks and paying a fortune for R&D. Imagine the value of a biotech that could raise its odds to 1 in 100. The majority of their activity could be running trials rather than doing more basic research.

That said, there is no clear way to invest in this thesis. Today's genetic medicine leaders may have locked themselves into dead-end paradigms. Some will certainly be acquired. Most will go bankrupt. Some will be 100-baggers. I suggest letting the big pharma companies do the diligence and buy them when they decide to acquire one of these smaller firms (their stock price will fall by the premium paid anyway).