
Lab testing per US FDA requirements for consumer injectables • Accuracy Matters. Results You Can Trust. • FDA #3034276914 • Women-Owned Small Business
Dorr, MI
Joined November 2024
- Tweets575
- Following91
- Followers1K
- Likes284
TrustPointe Analytics 🥼 retweeted
Replying to @mannyjplays
We have 100mg GHK-Cu in 6ml vials on their way to @TrustPointe right now. Big testing panel. You’ll be able to save yourself the hassle of moving into that second vial
TrustPointe Analytics 🥼 retweeted
Many thanks to @RegenRandy for taking the @TrustPointe ordering process for a spin and giving us some actionable feedback!
TrustPointe Analytics 🥼 retweeted
Replying to @TrustPointe
You set up an audit group? You utter madman, I love it.
TrustPointe Analytics 🥼 retweeted
Replying to @TrustPointe
This is exactly where the industry needs to go. Tested batch linked to the actual vial in your hand. No more taking a vendor's word for it.
TrustPointe Analytics 🥼 retweeted
Replying to @TrustPointe @huge_flood
Res Labs + Trust Pointe is goated. The best of both worlds
TrustPointe Analytics 🥼 retweeted
this is gonna ruffle some feathers but… subpar labs for testing peptides:
Freedom
ILS
BT Labs
Krause
great labs:
Chromate
Uzorak
Janoshik
TrustPointe
MZ Biolabs
TrustPointe Analytics 🥼 retweeted
Hot take:
People seeing the cost per vial from China are “outraged” by reseller pricing.
What they don’t see:
Spending all hours of the night on Whatsapp trying to find a legitimate supplier.
Hours of negotiation to arrive at pricing and a proforma invoice.
Taking a risk and wiring $20k overseas, knowing you may never see product.
If they ship it, there’s a 10% chance CBP deletes it at the border, and you have no recourse with the supplier.
If you do get it, you have to find a lab that tests it - and some are 2-3 weeks out. There’s a 20-30% chance it’s garbage. You are out the test money and the supplier will argue the results (or ghost you).
Then you sell it, and your processor fines you $50k for selling them. And people scam you by saying they never got the order.
It’s easy to see two numbers and not see all of the risk and losses between the two.
Prices will come down due to competition, but people are entitled to earn money for the risk and effort they put in.
TrustPointe Analytics 🥼 retweeted
We are very conservative on purity.
Unless we know what an atypical peak is, it is integrated and included in the purity value, causing it to be lower.
We have worked with willing manufacturers to inject multiple excipients on our methods to see if they align with the atypical peak - at which point we would be comfortable labeling that peak as an excipient and not including it in the purity calculation.
TrustPointe Analytics 🥼 retweeted
Replying to @RegenRandy @BrentPSorensen
I understand.
It’s tough because lab methods are IP, and the way a lab does something may be more efficient / better than other labs, and they risk giving up their sauce in a video. Ashlee won’t let me do one for that reason (she’s wiser than me).
This is handled in industry by requiring method validation data. If your lab has data demonstrating Accuracy, Precision, Linearity, Range, Specificity, etc according to a recognized standard like USP 1225, the proprietary method details on how they got there aren’t necessary.
Any lab attacking another lab better have method validation receipts.
Also remember that samples are rarely run in isolation, so if the rest of the samples on the HPLC run were fine, yours is probably just bad. We share the run data for every run, so systemic errors like a bad reference standard would be highly visible.
Re: 10mg vs 15mg errors
10 and it should have been 15
> sample mix up
> lost material during quantitative transfer
> incorrect sample dilution - overdilution
> sample not fully dissolved
> calculation error
> data entry error
> issue with reference standard (but it would bias all results on the run down)
> HPLC system/detector issue (system suitability checks should catch. Would affect other samples on the run)
> bad peak integration (cut off part of the peak, although it would have to be really bad here)
If they got 15 and it should have been 10
> sample mix up
> incorrect sample dilution - underdilution
> calculation error
> data entry error
> issue with reference standard (but it would bias all results up
> HPLC system/detector issue (system suitability checks should catch. Would affect other samples on the run)
> coelution due to poor separation (inflates peptide peak area)
> bad peak integration (integrated a huge front/tail and or other peaks including the peptide peak, although it would have to be really bad here)
Replying to @TrustPointe
I’ve used you guys a few times and was always happy with the service… I think accurate testing, fair pricing, quick turnaround times, and great customer service are key..
I got this from TrustPointe..
TrustPointe Analytics 🥼 retweeted
Well, USP <905> Content Uniformity on my RT10s from China FAILED
A 🧵
TrustPointe Analytics 🥼 retweeted
We left dry retatrutide in a truck and on a sunny windowsill all summer. The fridge group got the VIP treatment.
Then @TrustPointe tested the samples.
High purity. Lower measured content outside the fridge. Randy brought receipts. 🐀👇
peptidecritic.com/lab-report…
you’re paying for testing on the good ones too.
And you’re paying for the allocation of risk off yourself. You can legally buy research chems for yourself and not get in trouble at all. There is no fine for buying bpc 157 for research purposes for YOURSELF.
The RUO brands assume the risk for buying the chems and selling them under a thin guise and pass off the cost to the consumer for doing so.
The other alternative is to find a good Chinese plug and cut out the middle man but the downsides are very little of them do 3rd party testing and many have MOQs or “kits”
The allocation of risk and testing demands a premium that RUO vendors can pass off to consumers.
Simple economics really.
TrustPointe Analytics 🥼 retweeted
Replying to @RegenRandy @PharmerBobby
I guess my point is that the less expensive labs are going to get more expensive
Also there is an opportunity cost you are missing by not being able to sell product as your wait on testing
TrustPointe Analytics 🥼 retweeted
Replying to @JoshInProgress @TrustPointe
3-5 day turnaround is solid
TrustPointe Analytics 🥼 retweeted
Replying to @blankpeptides
heads up - while investigating another shop we found your freedom diagnostics COAs being reused with the client name changed to theirs and their product photo swapped in. the lab's search code still starts with "Blan" which is how it was caught. full writeup: peptigrity.com/discussions/h…
TrustPointe Analytics 🥼 retweeted
And we hadn’t even had a beer yet
TrustPointe Analytics 🥼 retweeted
Replying to @BowTiedUM
We have targeted lcmsms screens using 3 mrm transitions to identify 90+ illicit substances, method is validated for peptides, supplements and drinking water, and uses certified referance standards. We have screened more than 10,000 peptides and thus far none have been positive for any opioids, amphetamines, mdma, or related substances. We have also offered to test any presumptive positive results for free, but no one has sent in a vial yet. We have dozens of otc supplements that have tested positive for pd5 Dexamethasone, dicofenac, etc. We certainly have had peptides fail purity testing, or be labeled wrong, but the number pales compared to otc supplements that are currently (loosely) fda regulated.
TrustPointe Analytics 🥼 retweeted
Excited to talk about @reslabs_ai and what we're building.
So I started taking BPC back in 2023, then recently started buying vials back in Nov, and noticed at the start of this year how quickly the market was moving, vendors offering more advanced testing combined with growing consumer demand for first, endotoxins, then heavy metals, and also sterility now. However, I kept noticing a trust gap between, Lab/COA, Vendor, and vial.
The specific pain point I wanted to solve is the trust gap between the COA and the actual vial: a COA proves the batch was tested, but it does not prove the vial in your customer's hands came from that batch. I wanted to shout out here @RegenRandy for helping me think through how I could try and approach the pain point through a number of calls after I DM’d him a few months ago.
Our chips bind the COA to the physical vial so customers can cryptographically verify it with a tap on their phone (they don't need to download an app). It's a step beyond QR codes and is also the same approach big pharma has moved to for some high value medications. A counterfeiter can copy a URL or clone a QR code and point it somewhere else. But they cannot replicate the authentication code for the next tap, as the counter value changes on every interaction. (And also I’ve more recently seen some fake sties impersonating established brands, which our chips protect against too).
A bit about me: Back in 2019 I was working at @Consensys , on the Diligence team that audited smart contracts (it was a pretty new industry and in some ways also a 'gray market'). Hardly any project was doing audits at the time. Within a couple of moths it was the standard, and today no serious team launches without multiple audits.
That shift came from inside the industry, not from regulators. A handful of teams decided audits were the right thing to do. The rest of the market followed because the standard had been set from below.
My thesis behind Res Labs: The RUO market feels familiar for the same reasons. The same self-sovereignty culture that pushed against legacy banks is here too, this time pushing against big pharma. And the same dynamic plays out: if legitimate operators raise trust standards inside the market, it strengthens the whole category and gives good actors more power to shape the market on their terms, and ride the long term growth.
There's also a feeling I remember from those Consensys days. You can tell when you're inside something that's about to shift shape, and that's how this market feels similar. The infrastructure exists / is being built, the early movers are setting the new floor with incremental changes that each time push the industry forward in a positive direction (without having to wait for regulator daddy).
Yes, were still in pilot phase and so we are actively get real life operational feedback and improve the product / platform quickly. The goal is to collaborate with Vendors early on and shape this around what is best for you and your customers. Because we are still early, we are flexible on feature requests and can prioritize what matters most to your ops workflow and / or the product itself. Just let us know what you want and need from your side and we’ll do our best to accommodate.
Rome wasn’t built in a day.
Also want to shout out @ogsvg and @_9th_Life_ they have been a great resource with Reptides and they get zero from our business relationship (trust me ive tried lol I’ve offered to figure out some incentive structure and they have refused). They’re solid dudes and we have been on calls talking about Res Labs for the past month or so. And sidenote I use @reptidesco app now to help research and track my own stack.
If you have any Qs or wanna jump on a call DM me!
🤖 Made with AI
The future of the COA is here.
reptides.co/reslabs
No more COAs copied off a website. No more COAs swapped onto a different batch. No more wondering if the vial on your bench came from the lot that was actually tested.
Today we're proud to announce that Res Labs chose reptides as the exclusive partner to launch with.
Res Labs makes proof of testing tappable and verifiable, and impossible to copy or swap: a real lab COA, bound to the exact vial in your hand.
How it works:
1. A supplier tests a batch with a real lab, the same way they always have.
2. Res Labs binds that lab's COA to that exact batch and ships one tap tag per vial.
3. The supplier puts the tag on the bare glass and their own label on top.
4. You tap the vial with any phone. The compound, the lab that ran it, the COA, and a permanent on-chain record open right in your browser. No app, no account.
5. You can even see whether that chip has been tapped before you. We've spent the better part of the last month working with Res Labs behind the scenes, recruiting the founding suppliers for this first pilot cohort.
From here, the goal is to move verification to the very front of the supply chain: the manufacturer binding the COA to the vial at the source, before it ever reaches a supplier.
The earlier in the chain the proof starts, the less of the downstream risk any single supplier can control, or quietly change. reptides makes $0 from this.
Our job is to hold this market accountable, and binding a real COA to the real vial is exactly the standard we exist to push. reptides.co/reslabs
Suppliers who want to build the future: the cohort is open - apply at the link.
TrustPointe Analytics 🥼 retweeted
Replying to @TrustPointe
You can see posts about it on reddit.
They would buy the raw powder and then just weigh it into the vials.
No Lypholization, no sterilization and no vacuum sealing.
Hence the term, bathtub boys.





















