• Log in with Facebook Log in with Twitter Log In with Google      Sign In    
  • Create Account
  LongeCity
              Advocacy & Research for Unlimited Lifespans

Photo
* * * * * 1 votes

Supplement/diseases/Biochemistry pathway Network Visualization Software

software visualization

  • Please log in to reply
12 replies to this topic

#1 Justin BoBustinBananaFanaF

  • Guest
  • 30 posts
  • 0
  • Location:Here
  • NO

Posted 20 February 2019 - 08:15 PM


I am looking for software that will allow you to search for Supplement(s)/diseases/Biochemistry pathways.

If it doesn't exist I will build it, but I don't want to waste my time if it already exists.

I am also looking to gauge interest in such software. If other people would be willing to add to the tool would be exponentially more useful.

 

Example Use cases:

1) What does passionflower effect?

2) What biochemistry pathways are shared by Kava and Passionflower?

3) What decreases glutamate?

4) What increases the risk of cancer?

 

Each search would return a network graph and additional information would be provided for each node (it could expand the search to add focus to the node or it could provide research justify the claim). Eventually there would be a UI for up/down voting the relationship, adding new nodes and adding citations.

 

 



#2 Flux-Odyssey

  • Member
  • 3 posts
  • 2
  • Location:Genespace

Posted 21 February 2019 - 02:32 AM

The closest thing I can thing of are software programs that model metabolism in bacteria like E.coli and then allow you to manipulate the metabolic flux between nodes, knockout pathways, add new genes/proteins/reactions to see how it affects the system. It's tangential to what you want but here is Escher: https://escher.githu...-demo/knockout/

 

It gets more complicated for supplements because you have to go from supplement->active ingredient(s)->enzyme interaction->metabolites->more interactions?->downstream effects, like signaling cascades, increased seratonin -> long term effects like Long Term Potentiation or Hippocampus Neurogenesis. But I like complicated things, this is a cool idea. Do you have a background in software?



sponsored ad

  • Advert
Click HERE to rent this advertising spot for BRAIN HEALTH to support LongeCity (this will replace the google ad above).

#3 Justin BoBustinBananaFanaF

  • Topic Starter
  • Guest
  • 30 posts
  • 0
  • Location:Here
  • NO

Posted 21 February 2019 - 04:54 PM

It gets more complicated for supplements because you have to go from supplement->active ingredient(s)->enzyme interaction->metabolites->more interactions?->downstream effects, like signaling cascades, increased seratonin -> long term effects like Long Term Potentiation or Hippocampus Neurogenesis. But I like complicated things, this is a cool idea. Do you have a background in software?

 

What I am looking for would be significantly simpler than what it sounds like you are describing.

 

I want

1) a database that stores facts about the relationships between biological factors.

2) a way to visualize the information in 1

3) a way to update and search the information in 1

 

As a background I am a software developer with only the tiniest background in biochemisty.

I am coming at this as a bit of a research/organization nut.

I have sleep issues that I've spent 5 years trying to resolve and over the years I've searched this site, selfhacked, etc. for a variety of pathways and compounds.

 

I've organized my research by using a quadrant based system per biological factor (e.g. lifestyle modification,supplement,pathway,hormone,neurotransmitter).

For example, I have a piece of paper I've folded into fourths for Growth Hormone.

One corner of the paper is things that increase growth hormone, another corner is things that decrease growth hormone, another corner is things that growth hormone decreases, and the last corner is things that growth hormone increases.

 

I have hundreds of these papers (one per biological factor) that contain thousands of relationship facts.

I plan on loading those into a database and then enabling other people to add their own relationships.

 

Thank you for your interest/help



#4 ta5

  • Guest
  • 845 posts
  • 250
  • Location: 

Posted 22 February 2019 - 02:08 AM

I think such a site would become obsolete fairly soon. The thing to do is extract the relationships from the studies automatically using a natural language parser and what not. People are working on this. For example: Extracting Medical Information Using Machine Reading and the PDF linked at the top. NCBI has their UMLS tools which includes a Semantic Network and maybe it's a step towards a knowledge base, but it doesn't appear to be exactly what we are looking for. I'd be surprised if there's not something out there already. Maybe there is, but it's not publicly available.

 

It's tempting to take one of the open source NLPs and throw a bunch of abstracts at it to extract the semantic triples, and see what it finds.


  • like x 1

#5 Mind

  • Life Member, Moderator, Secretary
  • 16,209 posts
  • 2,000
  • Location:Wausau, WI

Posted 25 February 2019 - 04:46 PM

This is a good idea. I brought it up a few years ago. The problem is getting coders and other people to get it up and running. It is very difficult to get people to volunteer their time. With the right funding you could really get things done.


Edited by Mind, 25 March 2019 - 06:58 PM.


#6 cat-nips

  • Guest
  • 210 posts
  • 7
  • Location:Central Jersey

Posted 27 February 2019 - 05:29 AM

Noble undertaking.  Something along the lines for what treato used to be before they shut down in terms of analytics with an additional biochem pathway component, and a corresponding info/article section linked to a combined database of both professional and user-generated data to generate those analytics and content.  Pretty cool if you could do it.  :)


  • like x 1

#7 Justin BoBustinBananaFanaF

  • Topic Starter
  • Guest
  • 30 posts
  • 0
  • Location:Here
  • NO

Posted 28 February 2019 - 10:15 PM

I think such a site would become obsolete fairly soon. The thing to do is extract the relationships from the studies automatically using a natural language parser and what not. People are working on this. For example: Extracting Medical Information Using Machine Reading and the PDF linked at the top. NCBI has their UMLS tools which includes a Semantic Network and maybe it's a step towards a knowledge base, but it doesn't appear to be exactly what we are looking for. I'd be surprised if there's not something out there already. Maybe there is, but it's not publicly available.

 

It's tempting to take one of the open source NLPs and throw a bunch of abstracts at it to extract the semantic triples, and see what it finds.

 

This software would be free (with a plan to have some minimal adverts for support if at some point I have 1000s of users which would be somewhat unlikely) and mainly made for me as it is unlikely for any particular application to make it big. If at some point it is able to generate ad revenue enough for me to do it full time I'd love to look into additional data extraction.

 

Based on

https://www.nlm.nih....bases/umls.html

https://www.nlm.nih.gov/research/umls/

 

UMLS may be of value. I'd still want to have the ability to add whatever data I want so I anticipate even if I can figure this out it won't be enough.

 

 

I too would be surprised if something like this doesn't exist, but I can't find it. I'd assume that for legal reasons (and the amount of work it would take to do something complicated like what you are proposing) it wouldn't be free and also wouldn't allow me (or anyone else) to add my own relationships. One of the primary reasons for this thread is to see if it does exist and I'd love it if someone could provide one.

The site would be socially driven much like this one and would be anticipated to have a similar level of misinformation if it gets a decent number of users.

I don't expect rapid obsolescence but have no idea how to test this theory.

My feeling is that the relationship between some core things like MAO and histamine is likely to change. While I do think that new studies might come in that question Rosemarinic Acid's relationship to GABA the old studies/relationships may still have some merit. My main weakness will be for new information and if it is useful enough for me I will I add it manually.

 

I personally don't have much of a desire to discover new relationships using NLP, etc. Though if it doesn't render the visualization to slow or fill it with less useful information I'd be game for letting others do so by sharing the code on GitHub. On the other hand, I would like to as a phase to use PubMed's API (or another one if I can find an easier to use one) to link studies to relationships. I anticipate mostly manually reviewing the studies as to whether they support or refute the claim but at some point I might add some simple heuristics.


Edited by Justin Reiser, 28 February 2019 - 10:22 PM.


#8 adamh

  • Guest
  • 706 posts
  • 60

Posted 01 March 2019 - 02:20 AM

I'm just coming from left field with this so feel free to disregard if its been looked into and found unworkable. Why not simply make or modify an ai program to search through a database consisting of things like pubmed articles. There could be some false leads and faulty studies in there, no doubt will be some but it eliminates a lot of n=1 anecdotes which could be placebo or nocebo. 

 

You would want it to have access to a huge database which would be a hurdle to start with. Next you want it to be able to evaluate the worth of a study looking at factors like was it replicated etc. You would want the program to be able to find correlation and relevant facts. Lets say you wanted to look into rosemarinic acid's effects on gaba. It would of course look at all papers on ra and how it works. It would look into the metabolic pathways its known to affect and look for papers that look into that even those that don't mention ra of course and look for things that affect gaba. 

 

The output you might get could be all the known data pertaining to your search. It will hopefully discover some under appreciated or overlooked correlations. It would in effect strain like a smart sieve and discard irrelevant data and give you in a nutshell the pertinent info. It would probably not answer the question you had originally which might be for example how can ra help with sleep, with cancer, or etc but it would save countless hours of time looking at all those studies and data and hopefully have some serendipitous discoveries in it for the discerning eye to spot. There is no substitute, at present, for human intuition and judgement but it would do the leg work, the dirty work so to speak for you and for anyone who used it. It could also be worth a lot of money if you got it to that point and I don't mean just advertising.

 

Did I merely rephrase your search in slightly different terms? Hopefully this will make you look at it from a slightly different angle and perhaps help. If not, then it just took a couple minutes to read. 



#9 Justin BoBustinBananaFanaF

  • Topic Starter
  • Guest
  • 30 posts
  • 0
  • Location:Here
  • NO

Posted 01 March 2019 - 03:45 PM

I'm just coming from left field with this so feel free to disregard if its been looked into and found unworkable. Why not simply make or modify an ai program to search through a database consisting of things like pubmed articles. There could be some false leads and faulty studies in there, no doubt will be some but it eliminates a lot of n=1 anecdotes which could be placebo or nocebo. 

 

You would want it to have access to a huge database which would be a hurdle to start with. Next you want it to be able to evaluate the worth of a study looking at factors like was it replicated etc. You would want the program to be able to find correlation and relevant facts. Lets say you wanted to look into rosemarinic acid's effects on gaba. It would of course look at all papers on ra and how it works. It would look into the metabolic pathways its known to affect and look for papers that look into that even those that don't mention ra of course and look for things that affect gaba. 

 

The output you might get could be all the known data pertaining to your search. It will hopefully discover some under appreciated or overlooked correlations. It would in effect strain like a smart sieve and discard irrelevant data and give you in a nutshell the pertinent info. It would probably not answer the question you had originally which might be for example how can ra help with sleep, with cancer, or etc but it would save countless hours of time looking at all those studies and data and hopefully have some serendipitous discoveries in it for the discerning eye to spot. There is no substitute, at present, for human intuition and judgement but it would do the leg work, the dirty work so to speak for you and for anyone who used it. It could also be worth a lot of money if you got it to that point and I don't mean just advertising.

 

Did I merely rephrase your search in slightly different terms? Hopefully this will make you look at it from a slightly different angle and perhaps help. If not, then it just took a couple minutes to read. 

 

Thank you for the question. If I understand it correctly it has the same high level goal as ta5's excellent point.

I will try to clarify my intentions/ideas here. If I don't fully address your comment please see my previous response to ta5.

 

The problem I am trying to address is how to understand/manage the large quantity of health data (expressed in relationships between exogenous compounds and endogenous transmitters,etc.) I have accumulated. I feel the best way to solve that is with a visualization/search tool. A good Data Extraction tool if it exists could replace the need for my tool if it provides a good enough UI. If it doesn't provide a good UI or doesn't exist it isn't relevant to me at this point as it would be overkill for my problem.

 

The other factor is level of effort/time necessary to write the software. Data Extraction is much more time consuming than visualization/searching/managing and in my opinion is better suited to a team effort.



#10 Justin BoBustinBananaFanaF

  • Topic Starter
  • Guest
  • 30 posts
  • 0
  • Location:Here
  • NO

Posted 01 March 2019 - 03:52 PM

This is a good idea. I brought it up a few years ago. Thee problem is getting coders and other people to get it up and running. It is very difficult to get people to volunteer their time. With the right funding you could really get things done.

 

My solution is simple enough that it could be done by one developer in 1-3 months of full timeish work. It's also useful enough that eventually a coder will have enough time/health problems to accomplish it.

 

On the other hand, It might be a good idea to come up with a central location/format to store "health facts". If enough people were interested they could fill out a spreadsheet with relationship data that I could incorporate into my program or if my program falls through could be used by the next coder. I could provide a spreadsheet template if there is interest
 



#11 adamh

  • Guest
  • 706 posts
  • 60

Posted 02 March 2019 - 04:50 PM

A search tool is fine but what database is it going to be looking in? Are you talking about something like google or duckduckgo but with a focus on biochemistry? How would your search tool be better than simply using them? You are going to end up with a huge pile of semi-related studies and anecdotes. The relationships between various factors will not stand out until you painstakingly analyze the results

 

Thats why I say some form of ai is going to be needed. It will do the hard work of finding relationships between various drugs, various factors in the cell and so on. If your search tool simply uses the net, it will come up with a lot of junk. I think you can pay an annual fee to see all pubmed articles free can't you? If so that will help but if your search gives you 1000 results in order of relevance, you have a big job ahead of you to sift through it to find what you want. If ai is too expensive/ time consuming to do and search is easier, that may be the place to start. Later perhaps fine tune it and give it some ai capabilities.



#12 Justin BoBustinBananaFanaF

  • Topic Starter
  • Guest
  • 30 posts
  • 0
  • Location:Here
  • NO

Posted 08 March 2019 - 04:38 PM

A search tool is fine but what database is it going to be looking in? Are you talking about something like google or duckduckgo but with a focus on biochemistry? How would your search tool be better than simply using them? You are going to end up with a huge pile of semi-related studies and anecdotes. The relationships between various factors will not stand out until you painstakingly analyze the results

 

Thats why I say some form of ai is going to be needed. It will do the hard work of finding relationships between various drugs, various factors in the cell and so on. If your search tool simply uses the net, it will come up with a lot of junk. I think you can pay an annual fee to see all pubmed articles free can't you? If so that will help but if your search gives you 1000 results in order of relevance, you have a big job ahead of you to sift through it to find what you want. If ai is too expensive/ time consuming to do and search is easier, that may be the place to start. Later perhaps fine tune it and give it some ai capabilities.

 

The search tool will build its own database using user provided data. It will not search the net nor will it do any other sort of data extraction at least initially.

I will bootstrap it with data I've accumulated and users can add their own data or up/down vote existing data. Users could also add studies that confirm/refute relationships other people (including me) have defined.

 

There are a few main advantages to the search tool. There are disadvantages of course and many of them have been pointed out already (mostly by you, thank you for the good questions) .I don't expect the tool to be perfect but I see a niche for it. PubMed, Longecity, Google, etc. will still have their place. I opened this thread to see if the tool is useful enough for people besides me to use it

 

Here are 8 positives to the tool which might help explain why I want to create it and why people might want to use it

 

1) It will be a network graph.

 

It will be something like a much broader, dynamic version of a static network graph like https://www.google.c...iact=mrc&uact=8

 

It will show the complex series of relationships involved in chemistry much better than looking at studies one by one can.

 

A google search on Huperzine A will show that it decreases AChE and may even show that it increases ACh but it will not show

A) how ACh increases other things (like growth hormone).

B) show what else increases ACh (like sauna use)

C) show what decreases ACh (like CBD)

 

2) It will be focused.

 

2A) There will be one relationship per pair of objects.

 

Plain old googling will have many leads which will be hard to understand without a systematic representation of the data

 

2B) It will be focused on what its users care about

 

I threw out a bunch of my data as I learned what things are less relevant to me. I don't particularly care about motility at this point and knowing lemon balm's affect on it will merely clutter the data I care about. Of course if I ever get numerous users they may add motility relationships, but the tool will reflect actionable data to its userbase.

 

3) It will include anecdotes.

 

Just because there is no study for something doesn't mean it isn't sometimes true (and vice versa). For example, the tool might have a relationship between lemon balm and sleep which shows lemon balm increasing sleep as supported by https://www.ncbi.nlm...les/PMC3230760/ . However, I personally have a hard time sleeping after consuming lemon balm presumably since it is an AChE inhibitor. So while there would be the sleep/lemon balm relationship in the tool there would also be a lemon balm/AChE relationship which would help some users make more educated decisions

 

Personally, the two people who say that a supplement gives them energy are very valuable to me. I personally do rather poorly on most energetic supplements so rightly or wrongly this information lets me prioritize which strategy to try first. This anecdote will be especially useful if it is backed up by a mechanism (such as the supplement increasing NE)

 

4) It will be user driven. Mistakes will hopefully be hightlighted and perhaps eventually automatically removed.

 

5) It will be dynamic

 

Since there is too much data to show at once clicking on a node in the graph will expand the breadth of the data visualized. So if you search for huperzine A it will display one or two levels of relationships. Assuming I default the tool to 2 levels of visualization it would show that huperzine A decreases AChE which increases ACh. If you clicked on ACh you would see one what else influences ACh and what ACh influences. Clicking the ACh node again would hide this information

 

6) It will eventually be configurable

 

If there are enough users I will add logins so that we can learn about what the user wants. For example, if you don't care about motility you could click on the motility concept and tell it to permanently hide motility relationships. On the other hand, if you like me are affected by dopamine, NE, and ACh these relationships could automatically be highlighted and expanded

 

7) It will provide relationships between exogenous compounds

 

A search for lemon balm would show what classes it falls into. Besides showing that it is an AChE inhibitor and allowing you to see other AChE inhibitors it will list its common components. For example, you could see that lemon balm contains rosmarinic acid and then see what rosmarinic acid influences and what other exogenous compounds have rosmarinic acid. I seem to poorly respond to Rosemary (another rosmarinic acid source) and there is some mechanism to explain why I might do so. While this certainly isn't hard science it does help rule out potentially harmful treatments or at least helps to prioritize them.

 

8) The tool will highlight and attempt to resolve contradictory data/studies

 

I plan to have something like a green circle with a number representing the number of positive studies and a red circle providing the number of contradictory studies. It will also have thumbs up and thumbs down features like longecity.

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

 

I agree that semi related studies will be a problem. I am sourcing my initial relationships from this site, selfhacked and mybiohack. I have seen things in all of them that seem wrong or that at least I can't understand. To remedy this, I plan on including whatever study information I can find and I will eventually try to rank the relationship's quality based on this data. For example, was a study in mice and was it in vivo? If I don't link to a study other users could always do so.

 

I also agree that some AI will be useful. I personally don't care about every finding out every relationship between every possible system/compound, but I would like to

1) Test existing relationships by validating them with AI detected studies

2) Find new relationships between existing end points. For example, if Schisandra and sleep are two things users care about I would like to know the relationship between the two. If none of my users care about decreasing baldness then I will not care about what Schisandra does about baldness.

 

I feel an organic slow development will be the only way to go here. If I don't get additional development support or tons of users I will probably only add the features I need and add them as I need them or have time to do so.



sponsored ad

  • Advert
Click HERE to rent this advertising spot for BRAIN HEALTH to support LongeCity (this will replace the google ad above).

#13 BrainBoxer

  • Guest
  • 3 posts
  • 2
  • Location:SCOTLAND
  • NO

Posted 25 March 2019 - 01:26 PM

Much needed project.  At the moment its a free for all with everybody having personal interpretation and health plans.  Both visual data mapping and common data collection would be nice. At the moment I have to draw maps on paper. With insomnia I rely a lot upon logging all my interventions so that if my sleep really improves I can backtrack over them and try to untangle which works.   I use an approach which includes Biological, Psychological and environmental-social factors.  And this is why software needs to be as open as possible.  I have flowchart entity relationship mapper called Yed if I need to draw diagrams, be nice if it could have values loaded into it.

 

https://www.yworks.com/products/yed

 

If you are not aware, there are loads of biochemical pathway mappers out there for labs and researchers to build data that covers genes and major systems of the body and mind.  Drop us a line if you want links to these.  However the problem is they are tailored for research mapping purposes and not setup for personal healthplans, but you might find useful source code to adapt.

 

The route to get people involved is probably going to be through common data collection with online participation from facebook groups to a particular aim thats quite popular with life extentionists.   Such as to log all the data for people using metformin using a table.  Because its kind of a zero sum game at the moment with the every person for themselves.. so there has to be a common goal to overcome this before we can engage participation with mapping processes.

 

And just to second what some of the above posters were saying about Ai. Because we are all trying so many things at once its hard to untangle cause and effect where as deep learning could tease out the relationships.  However this is a bit impractical without a sizeable coding team.  But what is more attainable is to link to simple online free statistical tools to load information from group data in tables to test hypothesis being tested in terms of p-values.  Easy enough for most to use, then from there some might explore multivariable analysis.   

 

 


Edited by BrainBoxer, 25 March 2019 - 01:31 PM.






Also tagged with one or more of these keywords: software, visualization

0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users