Thread by @Biotech2k1

Taking a look at the #AI/#ML use in biotech:

This is all about using tech to lower the costs and improve the rates of success in drug development. We still see 90% of drugs that enter the clinic fail to reach commercial. They can cost over $1 billion do develop.

1/ The first company is $EVO. They have a suite of technology they use to help other companies develop drugs. They use both Transcriptomics and Proteomics along with AI and ML. They work with many partners and get a small royalty on the sales of products they help develop.

2/ I don't own them as I want companies with big potential. Getting just a small royalty isn't going to get a huge growth potential unless they have thousands of partners. This makes them lower risk, but also lower reward. That is good if you like that kind of company.

3/ I also think their valuation is really expensive even after the biotech crash. They are expected to earn around $750 million in revenues in 2022 and trade at over $5 billion market cap. That is upward of 2x the valuation of many other companies with their level of revenues.

4/ The second company is $ABCL. They use AI and ML to sequence and develop antibodies. They license their technology to other companies for use in drug discovery. They don't have any wholly owned programs for their own.

5/ That means their potential is much lower unless they develop many partnerships. They only get royalties on all the products they help develop. That is fine if you are looking for a safer company with many partnerships and don't mind giving up the big reward.

6/ I still think this company is really expensive at $2.7 billion with no commercial products and in such early development.

7/ My third company and top pick for myself is $RLAY. They use Computer Aided Design (CAD), AI and ML to develop drugs that target proteins and enzymes that play key roles in cell development and drive oncology.

8/ They have 1 program partnered with SNY and 2 programs in development that are wholly owned. They have massive potential for their mutant specific PI3Ka inhibitors. They could do $3 billon or more in that space over time.

9/ I think the company is expensive at $3.1 billion for only being in phase 1. I honestly have been trying to build this position since the IPO at the cheapest possible price, but it hasn't been easy. SoftBank owns over 25%, and they won't sell.

10/ My fourth pick and 2nd favorite is $EXAI. They have 2 programs. The first uses a single cell genetics to help improve oncology therapies. They also use all the available clinical and science data along with AI and ML to develop new drugs.

11/ They have a lot of partnerships with 3 big pharma partners. They have multiple small cap partners they share 50/50 development with. Then they have a few programs that are mostly or wholly owned. This gives them a lot of potential.

12/ I think the company is slightly expensive for such an early stage of development at $1.6 billion market cap. Part of that might have to do with still being in the lock up period for another week.

13/ My next company is $SDGR. They are using CAD, ML and AI to develop drugs. They license some of their software to $RLAY. They make most of their money from licensing out their software. Its not a huge money maker, but it helps pay for their own pipeline.

14/ They have multiple drugs in development that target proteins and enzymes that drive cellular growth in oncology. They are just moving into the clinic so they don't have much data to validate their science.

15/ At $2.4 billion market cap, you are basically buying the tech company with its software sales and getting the biotech pipeline for free.

16/ My final pick is $RXRX. They are using automation to do millions of experiment every week. They record all that data into their super computer. They use AI and ML to take all that data and develop drug targets.

17/ This is still in its early stage. Most of their pipeline is older drugs that were found the old fashioned way. They are the most advanced clinically with 3 programs in phase 2. There is a lot of potential in their indications if they work out.

18/ I think they are the cheapest of all the companies in this space at only $1.15 billion market cap. Half of their market cap is now cash. I would say expectations are very low. Even if one of these phase 2 programs work, it could be worth 3x this value.