ChatGPT: Practical AI revolution is here... (1)

The text below is a private opinion not linked in any form to my current employee.  

So, all hands on deck… the AI revolution everyone was talking about for a long time arrived. I played a bit with ChatGPT for some of the general R&D and chemoinformatics tasks… and saying that impressed is nothing, I am shocked by the quality of the generated text and code, and the future possibilities of those products.

Some people claim that ChatGPT will not replace search engines. With a bit of time, it will likely be a merge between personal assistance and worldwide web search engines.

ChatGPT and its successors will revolutionize how software is built, increasing productivity while reducing costs. That, in turn, gives programmers more flexibility and time for more complex tasks. Furthermore, these tools can help bridge gaps between different programming languages, making it easier for coders to operate in heterogeneous environments.

ChatGPT-like models will overtake enterprise text mining from the analysis of the Slack conversation to tech documentation and code analysis. This will allow enterprises to quickly understand the context of discussions, documents, and data and better target their decisions. This will reduce time spent on mundane tasks and allow more time for problem-solving and strategy development. So, from "Can you develop a disease hypothesis about geneX using all current internal and public data? (And don't forget to send regular updates)" to "What would be the best Halloween theme this year?"

It is crucial to create regulations around data privacy and how models will be used and monitored for accuracy. A legal clause prohibiting AI from using text or code for model training will be a norm in the future. It likely will not stop companies from using open information from model training…

It's a huge industry that will grow. With a subscription of 50 USD a month, it will generate 600 USD per subscriber a year; even with 100K subscribers, it will give 60M a year of cash flow. With 4.4 million software devs in the US and 12M users of JetBrain software worldwide, the number of subscribers will be way more.

Below are examples of routine tasks that popped into my mind during ChatGPT testing. My request is in bold with a comment, and ChatGPT answer is in italic. I played with a more complex request like "Link disease X with gene Y" or "How X regulates Y", and all provided answers were making clear sense, is not yet to the level of abstraction of an advanced scholar, but it will be there soon. 

Can you please list all drugs that have a carboxyl group? Ok, “all” was overkill but it did provide a list of 15 drugs that indeed have a carboxyl group.

1. Aspirin (acetylsalicylic acid)
2. Ibuprofen
3. Naproxen
4. Diclofenac

Drugs approved in 2022. Ok, that’s clear.

I'm sorry, I am an AI model that was last trained on new data in October 2021, so I do not have information on drugs that have been approved in 2022. However, you can check the FDA website or other reliable sources for the most recent information on drug approvals.

Drugs approved in 2021. That’s perfect... the list goes on… 

1. Veklury (remdesivir) - an antiviral drug for COVID-19 treatment.
2. Vaxzevria (AstraZeneca COVID-19 vaccine) - a vaccine for COVID-19.
3. BNT162b2 (Pfizer-BioNTech COVID-19 vaccine) - a vaccine for COVID-19.
4. Moderna COVID-19 Vaccine - a vaccine for COVID-19.

What is the most interesting and unusual drug approved in 2021 and why? That’s a perfect answer. Tafamidis is the first non-PET compound for IDP-driven diseases. I would probably choose between tafamidis – a transthyretin stabilizer, sotorasib - a first-in-class KRAS compound and inclisiran – a siRNA inhibiting translation of PCSK9.

One of the most interesting and unusual drugs approved in 2021 is Vyndaqel (tafamidis) which is used to treat transthyretin amyloid cardiomyopathy. This drug is considered unusual because it works by stabilizing a specific protein, transthyretin, that is involved in the formation of amyloid deposits in the heart. Transthyretin amyloid cardiomyopathy is a rare, inherited condition that leads to progressive heart failure. This drug is considered interesting because it is the first treatment to specifically target the underlying cause of this condition and it is also the first treatment that has been shown to slow the progression of the disease. It's also interesting because it is one of the few drugs that have been approved specifically for a rare disease.

What are the main limitations on the application of the PROTACs for disease treatment? Nothing to comment, a perfect list.

Proteolysis-Targeting Chimera (PROTACs) are a class of molecules that can be used to target specific proteins for degradation in order to treat a wide range of diseases, including cancer and neurodegenerative disorders. However, there are several limitations to the application of PROTACs as a disease treatment.

1. Target specificity: One of the main limitations of PROTACs is that they may not be specific enough in targeting the intended protein. This can lead to off-target effects, which can cause toxicity or other unwanted side effects.
2. E3 ubiquitin ligase specificity: PROTACs require specific E3 ubiquitin ligase enzymes for their activity, which makes it difficult to predict the efficacy of the treatment in different cell types or organisms.
3. Drug delivery: PROTACs are relatively large molecules, which makes it difficult to deliver them to the target cells or tissues in the body. This can limit their efficacy and make it difficult to use them as a treatment for certain diseases.
4. Cellular uptake: PROTACs are often not taken up into cells effectively, which can limit their efficacy.
5. Cost: PROTACs are relatively complex molecules to produce, which can make them expensive to develop and manufacture, which may limit the accessibility of these drugs.
6. Resistance: As with any drug, there is a possibility that the targeted proteins may develop resistance over time, which could limit the effectiveness of the treatment.

Despite these limitations, PROTACs are still considered a promising avenue for the treatment of a wide range of diseases, and ongoing research is focused on developing strategies to overcome these limitations.

How to improve cellular update for PROTAC compound? ChatGPT can perfectly replace deep literature digging and save a lot of time.

There are several strategies that can be used to improve cellular uptake of PROTAC compounds:

1. Lipidation: By attaching a lipophilic tail to the PROTAC compound, it can be made more hydrophobic and thus more likely to cross the cell membrane.
2. Peptide modification: By attaching a cell-penetrating peptide, such as TAT or penetratin, to the PROTAC compound, it can be made more able to cross the cell membrane.
3. Conjugation with a cell-specific targeting molecule: By conjugating the PROTAC compound with a cell-specific targeting molecule, such as an antibody or aptamer, it can be directed to specific cells.
4. Chemical modification: The PROTAC compound can be chemically modified to make it more hydrophilic, which may enhance its uptake by cells.
5. Co-treatment with endocytic agents: By treating cells with agents that promote endocytosis, such as cholesterol, it may be possible to improve the uptake of the PROTAC compound.
6. Using the cell-penetrating peptides (CPPs) : CPPs have been used to enhance the delivery of various therapeutics into cells by facilitating their passage through the plasma membrane.

It is worth noting that these strategies are not mutually exclusive, and combination of them may be more effective in enhancing the cellular uptake of PROTAC compounds. Additionally, each strategy should be optimized for the specific PROTAC compound and the targeted cells.