I would like to find out whether the mixed protein extract of phalaris would convert tryptophan at a useful rate. These enzymes are active and are known to produce DMT when isolated from the plant cells since this is how their activity and kinetics has been evaluated in studies.

Phalaris seedlings can be mulched, the cell membranes broken, and juice filtered off.
The proteins in the juice could be precipitated by ammonium sulfate precipitation or filtered out with dialysis tubing (they are in the dalton range of common and cheap tubes) or both to separate them from the small molecules (this is a simpler process than it sounds). Filtering with different dialysis tubes or doing a proper fractional precipitation would eliminate a lot more of the unwanted proteins.

This extract would contain the enzymes of the DMT production pathway along with other enzymes and proteins, it would not contain gramine or such things.
The enzymes must be fed with tryptophan and SAM-e which are common nutritional suppliments.
The enzymes can be reused when you filter them out again with dialysis tubing.
The used SAM-e products are the same as in human systems but should be easily removable as they are highly water soluble, even if they get broken down further by remaining SAM-e cycle enzymes.

So you should end up with a solution that you can simply acidify a little and extract into organic solvent, then head straight to crystallisation. The main impurities would be the tryptophan to DMT intermediates which are all fine and are hopefully mostly used up anyway.

Here is a paper that I got the most relevant information from, it has details on the three enzymes in phalaris and their behavior:
https://www.scribd.com/document/43750195/N-N-Dimethyltryptamine-Production-in-Phalaris-aquatica-Seedlings
My biochemistry knowledge does not extend far enough to understand the reaction rates and binding coefficients and such yet but this seems as though it could likely be practical, and a lot nicer than buying INMT enzyme from rabbit lung extract.

One possible problem could be if it were far too slow once the products built up (they compete with the input molecules for binding on the enzymes) although if the enzymes are not attacked by anything you could simply leave it sitting for a month. From my understanding of the rates of reaction this should not take anywhere near that long with a achievable amount of enzyme. Another possible problem is if there are other enzymes that break down either the DMT pathway enzymes or the DMT itself that cannot be removed easily but I am guessing this is unlikely as most such enzymes need more than one input and only the tryptophan and SAM would be supplied.

Swapping the tryptophan for serotonin should yield bufotenine and 5-Methoxytryptamine should yield 5MeO-DMT but the enzymes might be more specific to tryptamine than the INMT enzyme which I presume they are similar to.

Sorry for plopping down an idea-dump for my first post here but this seemed like the best place to get some help with this and if it shapes up to what I hope it could be then maybe even contribute something to the community. I found a few references to this but couldn't find much that went very deep into it, although there is a fair bit on the topic of the more attractive but substantially more expensive enzyme-producing GMO idea.

I can't believe nobody has responded to this yet.

I know enough about the subject to see that it might work and that's about it. I speculate that there are some missing details that could potentially make this process slow and complicated. I hope that I'm wrong but that's what I see.

What I would really like to see is selective breeding, hybridization, or even genetic modification that could lead to phalaris that could be used in ayahuasca and simple extraction teks such as is used for MHRB.

Yes I was a little disappointed too but it is a bit far out there so scepticism is quite understandable.

Breeding would be a much nicer option, both DMT and gramine expression is highly variable so it would likely be quite quick and easy to selectively breed it. Hybridization might even break the gramine pathway completely since P. uses different enzymes to produce it, although GM would be a much faster way to that if you had the lab for it.

If you planted gramine-free phalaris outdoors you would want to prevent it from seeding itself though since grass pollen travels so far, and producing seed for that would probably have to be done in a glasshouse which would be a bit of a bother but certainly worth it. The initial breeding could be done anywhere though since you are testing it. I would really look forward to this though since once it is done it would be very easy for folks to grow and extract without much difficulty or risk.

I did a bit more research and it looks like the conversion would not be fast enough to be completed in a day, but would certainly produce DMT faster than would occur in the same quantity of the phalaris you extracted the enzymes from. To counter the slow speed you could heat it up as far as it will go without cooking the enzymes, add some easily removable preservatives, and leave it sitting in a cupboard for a couple of weeks and probably end up with more DMT than you know what to do with.

The main thing that I think could be a source of complication is the preservatives that you may need for long incubation, many methods are available for this but hopefully the most basic methods or even doing nothing at all will suffice. I would need to find someone who knows a lot more to find out if I've overlooked something though and would need to actually do it to find out if there was something more subtle missing.