Phytochemical databases disappearing?

Question

[alchemizt33]

I'm looking for databases with information on the compounds present in plants. FooDB and Dr Dukes databases are brilliant. There used to be one called The Herbal Interaction Target Database, I tried searching it recently and see that it is no longer online. Luckily, I made a backup of the whole HITD database a few years ago so I can put it back up online.

Today I searched for phytochemical databases, I found a page which lists all of these:

Ecosys--plant ecological ranges

EthnobotDB--worldwide plant uses

FoodplantDB--Native American food plants

MPNADB--medicinal plants of Native America

PhytochemDB--plant chemicals

I tried following the links and every single one of them said "Server not found". That's just a few of them, there are loads of databases like this like:

http://www.mapsdatabase.com/

http://bioinform.info/

http://home.iitj.ac.in/%E2%88%BCbagler/webservers/Phytochemica

There are lots of good databases still available online, but it troubles me that so many databases are disappearing like this. I know that the HITD database contains amazingly useful information that I have never seen anywhere else, so it is troubling that a database like this just disappeared without a trace.

As computers evolve and things like AI and machine learning are emerging, our abilities to use these bioinformatics databases to do amazing things are growing to the point that whole new worlds are opening up in terms of what we can do with this data. I feel inspired to do something about this issue to databases disappearing, like start a website where you can access it all from one place through an API and will be permanently available to everyone.

I'm backing up all the existing databases which seem like fragile ones that might disappear in the future. But its the ones that have already disappeared that are on my mind, I need to find a way to get the data of the databases that have disappeared. Maybe backups exist somewhere on the internet. I tried finding a backup of HITD though and couldn't. I tried the wayback machine but it only has snapshots of the index page, not the actual plant pages.

Any ideas on what I can do?

[GenoMax]

so many databases are disappearing like this.

Not surprising. Funders are not keen on funding upkeep of databases. And since end-users are not very keen on a pay-for-use model, something has to give.

If you made copies of existing databases that are no longer accessible, you should be careful about putting them up for public access. You would want to obtain permission from real owners at a minimum to not find yourself in legal trouble.

Answer 1

[Biofind17]

Like you mentioned, computers and AI are progressing so much, almost all you need to do is get all the phytochemical names from those databases and Pubmed, who will always use AI well, will handle the rest. I took the Duke phythochemical list for this github test: https://github.com/GithubUser017/EntrezPlantSearch . The thing we really need is a database that will take a plant miRNA nucleotide sequence and predict cross-kingdom interaction with animal transcriptome. There was an interesting Japanese paper where they used polyethylene glycol to inject corn derived exosome like nanoparticles: https://pubmed.ncbi.nlm.nih.gov/36352200/ . I don't have institutional access to read the newer papers on the hurdles of this type of research: https://pubmed.ncbi.nlm.nih.gov/36947052/ . It seems like right now there is a lot of avoidable experimental trial and error. Scientists just try culturally significant or common ethnobotanical plants.