AI-Assisted Research

This is a place to post research you have done on the topic along with your conclusions.
themkal
Posts: 7
Joined: Tue Jan 24, 2023 1:40 am
Contact:

AI-Assisted Research

Unread post by themkal »

Dear All,

First post here and I will try to keep this as short as possible . I had Post Finasteride syndrome that turned to ME/CFS (Chronic Fatigue Syndrome). Since 2011 I have been using Machine Learning methods to try to put the pieces of the puzzle together. Being a Data Scientist since 2000, I couldn't think of a better way than using Machine Learning and other analytical methods to identify what is going on. In 2015, i started developing a software framework that would help me accelerate my research. I refrained from looking at PFS as I had ME/CFS and so I started researching it. In 2018 i filed for a patent for this methodology , which you can see here :

https://uspto.report/patent/app/20180322958

Before continuing : I wish to express that I do not ask for any compensation but I will need your help. And getting this help is easier said than done (more on this later)

I have evidence that the methodology I have been using outperformed (=found much earlier) several findings of researchers. Two examples below :

BH4 (Tetrahydrobiopterin), Tweet in 2019 (The new gene that AI identified was Sepiapterin - gene SPR)

https://twitter.com/lifeanalytics/statu ... 8693214209

..and study from Stanford in 2022.

https://studypages.com/s/study-to-inves ... id-525299/

Next, we have Peroxisomal Dysfunction. Please see the following post (posted in 2017) : http://algogenomics.blogspot.com/2017/0 ... p27a1.html

In the generated graph (using Network Analysis) you can see some nodes (among others) with names "peroxisomes", "pparalpha", "ppargamma" and "Choline deficiency"

5 years later we have the following study, where Peroxisomal dysfunction (and Choline deficiency) were identified : https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9320121/

These are just two examples. There are many more but I wil not get into this further, I am more than willing to send additional evidence upon request.

Ever since 2017, I have been trying to have research organisations and researchers to use my system. No one decided to give it a try apart two researchers at Stanford. However they have used the system for very specific tasks. A researcher with specialty in Immune system will look at the immune system. A researcher looking at metabolism will look into metabolism, etc. We need greater collaboration, not "research silos". I am just lucky enough to be again fully functional, having overcome ME/CFS for the past 7 years but I had no luck in having researchers use this system.

Regarding PSSD

In 2017 , I also sent to Professor Healy the Network Analysis graph which I mentioned. I told him that the main problem -according to the system- is the liver. Understandably, he asked me to tell him what is the mechanism. I explained to him that I couldn't help on that but I was willing to work with other researchers in order to do a better job and come up with specific solutions. No other steps were taken.

In December 2022, I was contacted by a PSSD patient (I will not disclose his identity, he can do it if he wishes) and suggested that I should talk with Professor Luisa Guerrini. Theis patient introduced me to her. I saw her work and noticed that she was working with a pathway I came across during my research, namely P53 (TP53)

Post from 2018 (observe P53 next to SP1 transcription factor) :

https://twitter.com/lifeanalytics/statu ... 53/photo/1

The day after christmas we had a 2 - hour Zoom call. I explained to her all the work I have been doing all of these years (and got an invitation to talk to University of Milano!) so we began to see how we could use the software. An idea I had was to identify genes related to certain medications (including SSRIs) and see whether we have some common genes (unfortunately I cannot upload a snapshot in the forum)

Then, this list of genes were fed to the AI system and the system responded with a number of research targets. Prof Guerrini has shown great interest but she has a job to do. It is not possible to stop what she is doing. Professor Healy is aware of the results but he also explained to me that this direction of research cannot be looked at this point.

So, I am asking you what are the next steps to take? I emailed Dr. Melcangi however I had no answer from him or any of his team members. It is not wise to start suggesting things to do or medications to take. We have to do research responsibly. Please do not send any messages asking for help.

I do understand , I have been there. But we have to do research in a responsible way. I look forward for your comments and help. Thank you.
Last edited by themkal on Wed Jan 25, 2023 1:39 pm, edited 1 time in total.
shadowMan
Posts: 80
Joined: Thu May 18, 2017 9:25 pm
Contact:

Re: AI-Assisted Research

Unread post by shadowMan »

I was the one who introduced themkal to Luisa, he need our help to do it the right way, honestly, is our best chance, first he too endured the same torture as us, so, he knows very well the pain! and he was able to get out of this mess! How many years are we theorizing things? chasing our own tails? NOTHING came out of it, Ive mentioned themkal work a couple times here... trust me, he is MUCH MUCH ahead of everything that we saw researchers doing! This isnt a simple receptor X or Y issue... it is systemic and there are elements who can vary from person to person (Correct me themkal if im wrong). But, as themkal said, it must be done in a responsible way.

As he said, he doesnt want any compensation, he needs help from researchers to lead this responsibly! So, any suggestions on how we can help?
Terabithia
Posts: 78
Joined: Sun Aug 21, 2022 4:06 am
Contact:

Re: AI-Assisted Research

Unread post by Terabithia »

I also think the liver is the root source of this condition. Would you be able to post the nodal network for PSSD?

I believe what you’re doing is revolutionary and that AI assisted medicine is the future. Is there some way you could open source the project on the PSSD front so that we can help you provide the inputs required for your machine learning to work better?

What is the best case scenario to help you proceed with this research?
themkal
Posts: 7
Joined: Tue Jan 24, 2023 1:40 am
Contact:

Re: AI-Assisted Research

Unread post by themkal »

Terabithia wrote: Wed Jan 25, 2023 1:12 pm I also think the liver is the root source of this condition. Would you be able to post the nodal network for PSSD?

I believe what you’re doing is revolutionary and that AI assisted medicine is the future. Is there some way you could open source the project on the PSSD front so that we can help you provide the inputs required for your machine learning to work better?

What is the best case scenario to help you proceed with this research?
Yes, AI changes everything. No single mind can put together so many factors that play a part on a complex biological organism.

To answer your questions along with some comments :

-I believe that there is at least a common biological mechanism that is disrupted among all of these syndromes (PFS, PAS, PSSD, MECFS and possibly LongCOVID)
-We then have a number of individual-specific issues that need to be addressed -unfortunately- in a personalized manner (my hypothesis)
-Our bodies have fallen to a vicious cycle. It is as if you want to ask from an engine to go with 80 MPH but all it can give is just 10.
Would you be able to post the nodal network for PSSD
The methods I use are not just nodal networks, I am using several techniques looking at the same problem. It is as if you have a number of experts (actually algorithms) that suggest targets and then all expert's opinions are averaged in order to identify common targets. But I could do it for PSSD, yes
Is there some way you could open source the project on the PSSD front so that we can help you provide the inputs required for your machine learning to work better
Please define "Open-source" as I am not able to understand. The way that the methodology works is described at the patent. Then all that is needed is me, the software framework and having "back and forth" information between the AI tool and medical experts. My work with Luisa, took literally 24 hours to run.

The best case scenario needs just one word : Collaboration. And unfortunately, this is very tough to implement for reasons that I hope we all understand. As I mentioned on Reddit, this is personal for me as well. My life was destroyed by PFS then ME/CFS. I am very commited but I cannot do this alone.
healingspell
Posts: 42
Joined: Mon Nov 14, 2022 5:25 pm
Contact:

Re: AI-Assisted Research

Unread post by healingspell »

That's great. Hope it works out.

It would be better not to suggest you might know some medication that would help though. Don't wave bread in the face of the starving. It's cruel.
shadowMan
Posts: 80
Joined: Thu May 18, 2017 9:25 pm
Contact:

Re: AI-Assisted Research

Unread post by shadowMan »

healingspell wrote: Wed Jan 25, 2023 5:45 pm That's great. Hope it works out.

It would be better not to suggest you might know some medication that would help though. Don't wave bread in the face of the starving. It's cruel.
Please understand, this is not a condition that one medication can sort, its systemic and as told before, unfortunately, there are individual components involved. These intricancies make this condition so hard to treat/cure even diagnose! Its not waving bread in the face of the starving, Im probably one of the most ill with the condition here, I want A LOT to get out of this, to worsen things out, my young brother is developing this shit too, and it kills me to see him in the same path as me... But as I said, due to the nature of the problem, its not that simple. I saw a little bit of what themkal did, I know too how the researcher was amazed seeing what he did(Its a shame that she is in the middle of a postdoc research right now, she would be the perfect one to help with this). But please, again, themkal needs help to do this.
themkal
Posts: 7
Joined: Tue Jan 24, 2023 1:40 am
Contact:

Re: AI-Assisted Research

Unread post by themkal »

healingspell wrote: Wed Jan 25, 2023 5:45 pm That's great. Hope it works out.

It would be better not to suggest you might know some medication that would help though. Don't wave bread in the face of the starving. It's cruel.
I am not here to sell false hopes. As I mentioned, I am more than happy to provide evidence that the system has identified certain problems years before other research efforts. What I am bringing to the table is the possibility that at least one common mechanism between these syndromes has been found. All I am asking from you is to help me find out if this is true. Not by telling you what to do to get well (I made that clear from my first post), but by evaluating these hypotheses in the proper, scientific way.

If you were me, what would you do? How would you proceed?
sylv
Posts: 200
Joined: Thu Apr 27, 2017 6:32 am
Contact:

Re: AI-Assisted Research

Unread post by sylv »

It would be easier for us to give you some input to your hypotheses if you share what associations and data have your software generated.

Without it I'm not sure what is the real intent behind your post and what you expect from us.

You can upload data to cloud based server and provide a link if the attachment on this broken forum doesn't work
themkal
Posts: 7
Joined: Tue Jan 24, 2023 1:40 am
Contact:

Re: AI-Assisted Research

Unread post by themkal »

sylv wrote: Thu Jan 26, 2023 1:52 am It would be easier for us to give you some input to your hypotheses if you share what associations and data have your software generated.

Without it I'm not sure what is the real intent behind your post and what you expect from us.

You can upload data to cloud based server and provide a link if the attachment on this broken forum doesn't work
This is not possible, because the next thing will be people trying various supplements on themselves. I am more than willing to share these results with researchers and organisations (which I already have with Luisa Guerrini and Dr. Healy)
lukejimmy
Posts: 60
Joined: Wed Aug 07, 2019 8:29 am
Contact:

Re: AI-Assisted Research

Unread post by lukejimmy »

Are you this guy?
https://forum.propeciahelp.com/t/resear ... ning/34456
Any intel on "Enterohepatic Circulation" secondarily Cholestasis?

You're right on the money if your mariovitali
Post Reply

Who is online

Users browsing this forum: Ahrefs [Bot] and 4 guests