0:00
Thanks for having me everyone and sticking it out until close to the end of the day.
0:03
I'm excited to talk to you about CDD Vaults and everything we offer.
0:06
The only thing you really need to know about me in this moment got my background in biology but I'm a prolific lover of baking.
0:14
So after the talk, please come talk to me about everything baking and the science too as well.
0:20
So the too long; didn't read for this presentation.
0:23
CDD Vault allows you to register your chemical and biological entities that you're synthesising.
0:28
Our architecture is fully FAIR compliant.
0:31
You can link your research data to those molecules that you're synthesising.
0:35
You can create plots, visualisations, charts directly within the software and link it to those entities.
0:41
We have an integrated ELN, which because everything's connected together, that gives you a really powerful search.
0:47
We also have a couple built in AI tools that I'm going to talk about at the end.
0:51
So where did CDD Vault emerge from?
0:53
Became one of the fact that modern workflows are becoming more and more complex.
0:57
Even if we take overly simplistic workflow here where you've got molecules that you want to synthesise, you're making them running some assays so that you can do some modelling, each of these is going to be generating their own unique file types, right?
1:11
And if you're repeating this cycle, a thousand, five thousand, ten thousand cycles to get enough compounds to do modelling or to get your hit, this is a huge amount of data to keep track of.
1:21
These are also in essentially data silos where if you're the sorry person moving information through this workflow manually, that's going to be a lot of time.
1:30
And it's also error prone, not even considering where are you putting all of your data?
1:35
You may have a lab storage solution.
1:38
And then you've maybe got personal splinters that go off of there.
1:41
Like, you know, if I've got my own data, I promise I'll sync it with the main labs, but doesn't actually happen and doesn't actually get over? Maybe not.
1:47
It's not considering that you've also probably got CROs that you're working with.
1:52
So where are all these files going?
1:55
Are they in the right place?
1:56
Do you have version control issues where one file is being used, used one another one is a more current file.
2:02
So that's where we come in.
2:04
But first, does anyone, I mean, show of hands, we've got a few people here.
2:07
Has anyone ever gotten lost in their workflow before or not really been able to find their data?
2:13
Perfect.
2:14
Couple hands, yeah, going up.
2:16
So that leads us to this great quote from William Pollard.
2:19
So information source of learning, but unless it's organised, processed and available to the right people and in a format that primes it for decision making, it's going to be a burden and not a benefit.
2:29
So that's where we come in.
2:30
We're your single source of truth during your workflow.
2:34
So all of your data is interoperable and linked to the chemical structures that you're actually synthesising, stored within our secure CDD Vault environment.
2:44
So it's safe, interoperable and accessible.
2:48
So how are we going to take care of this old workflow?
2:50
How are we going to fix this?
2:51
We're going to first take care of this disjointed record keeping and replace it with a chemically aware registration system so you can see through.
2:59
We can set this up really quickly.
3:01
So you're clicking through.
3:02
I don't want to be able to look at all this or see it, just to see that there is a lot of customisation that you can do within your CDD Vaults.
3:10
But what's great is that we can get this set up and configured within 90 minutes.
3:14
So some of our competitors, you'll spend weeks and weeks having to go back and forth.
3:18
They're building out your user interface.
3:20
Ours is plug and play and we can get it ready, and you can start using it, adding data to it in under 90 minutes.
3:27
It's also really easy to bring your legacy data into CDD Vault.
3:30
So we've got templates that you can build out, upload your small strings, your structured data files, bring that in, make those templates so you can access your legacy data too.
3:38
You can also draw your chemistry.
3:40
We've got Ketcher integration.
3:41
So this is going to behave very similarly to ChemDraw if you're familiar with that, because the software is fully chemically aware, we're not going to let you make duplicates.
3:51
So if the software's seen the molecule before, it's going to prompt you to make a new batch rather than adding a duplicate molecule.
4:00
I also want to mention I'm using small molecules, an example for this talk, but we can register a lot more than small molecules.
4:06
So we can, we've just added antibody drug conjugates, nucleotide sequences, macromolecules, you could even load cell lines into a CDD default.
4:15
And because the software is chemically aware that extends through into the biologics.
4:20
So you can have this simplified biologist view.
4:23
I wonder if I can use my clicker here so you can see you've got the high level information on the monomers for the cyclic peptide.
4:31
We can have modified monomers, but if you click over to the full atom view, that underlying chemistry is all still represented within the software.
4:40
So you could toggle between your biologist and your chemist views very easily.
4:45
This is best resolved with our antibody drug conjugates that we just rolled out.
4:49
So you can load in your heavy, your light chains and depending on the information that you have available, we can generate different loadouts.
4:58
So the image on the left is without the structural information for this antibody.
5:03
If that's not known, this is what you'll see.
5:04
But on the right, we have that information built in.
5:07
And then if you've got drug conjugates and you know the sites that those drug conjugates are bound, we can have that information all displayed.
5:14
So again, you're seeing a simplified view, but CDD Vault knows all the chemistry and the atom by atom structure of this entire ADC, which is something that our competitors don't quite do.
5:25
So now we've got this integrated framework where we can put all of the research data.
5:30
How are we going to collect it consistently?
5:32
That's where assays module comes in.
5:34
You can build out your protocol forms, protocol definitions, these are ontology based.
5:39
You can also build your own if you want to, to capture all of that rich metadata.
5:44
As you start to do your assays, it populates the property grid.
5:47
This behaves like a workflow.
5:49
So you're going to see all the potential assays on X here, the different categories on the Y.
5:54
So as you're doing things right, you're going to be able to see exactly where you are on your workflow.
5:58
And this is going to get you complete accessible records.
6:00
So what would have been disjointed tools, you know, Word documents, Excel files floating around the lab or were still that paper note with what I did on it that particular day, which kind of disappeared into the ether of the lab.
6:14
It's all in one place and now it's fully searchable.
6:17
So you might be saying, OK, is this really complete records?
6:20
I have unstructured data.
6:22
That's where our electronic lab notebook comes in.
6:24
We can do a lot of the things that you'd expect from an ELN tool, even though that's not where we started.
6:29
So you can build out objective purposes linked to previous synthetic steps.
6:33
A couple of things I really want to point out that differentiate our ELN you can load in Excel docs and Word docs.
6:39
These are fully integrated.
6:41
So as you work with them inside our ELN, they update wherever they're stored.
6:46
And the main thing that you really want to, if you're a small biotech and you're trying to find a software to capture this information, which you always want to ask as you're looking around, is can I get my data out of the resource?
7:00
So you can see we have the export entry wherever you go within CDD Vault what we're going to let you get your data out when you need to.
7:06
And we're not going to hold it hostage.
7:09
And so you'll find that throughout the software, you can also draw chemistry directly within the ELN and we do support parallel synthesis.
7:18
As you draw this out, we're going to populate your stoich table for you, give you automatic safety warnings, calculate all the properties you expect, give you expected yields, etcetera.
7:28
Where we really bridge the unstructured aspect is when you're writing out your procedures or what happened on that particular day, you can link the entry to molecules, structures registered within the CDD Vault so that the molecule knows exactly what ELN entries it's linked to and vice versa.
7:47
So we're kind of bringing bridging the structure and the unstructured data here.
7:52
And what this really does is enable a truly powerful search through the software.
7:57
What we can do is build out what protocols we want to look through, search for structures or substructure properties.
8:04
So we can pop open Ketcher and say I want to look at this particular substructure here.
8:09
And then it will load that in and then screen through the entire CDD Vault.
8:14
And everywhere that particular substructure occurs, it will find it.
8:17
Bring it into this search as well as all of the entity fields and then the assay data that we wanted to look for.
8:24
And then this can be saved and replicated again later on.
8:28
So now we've got searchable complete records.
8:31
I also want to mention that we do have an inventory module as well.
8:34
I'm not showing it today, but that's also fully searchable and lets’ you bridge your real world to your structured data as well.
8:40
So you know what you have.
8:42
And that has a lot of things you'd expect from an inventory module.
8:45
So now that we've got complete searchable records, we've collected all your assay data and put it in a framework.
8:51
How do we make it meaningful?
8:52
We want to do something with those results, right?
8:54
So we can start with our activities module.
8:57
This is where you're going to be able to create calculations and plots within those saved searches.
9:01
So we've got some raw data that's not being shown here, calculated into the IC50, the hill slope, and then we're plotting that.
9:10
But because this is a framework, when we drop in batch 2, these are automatically populated.
9:15
So rather than having to take our Excel, put it in Prism, make sure we use the right version, and all of this is going to be automatically populated for us, and we can create averages.
9:26
That was the Hill equation.
9:28
But we with our curves module, we have a number of fits that are available.
9:31
So we're constantly adding to these.
9:33
Some recent ones, enzyme kinetics, exponential accumulation, decay, PK intravenous, PK oral dosing, these can all be accessed and plotted within that safe search.
9:42
So instead of having to pop open Prism and you know, make sure you're using the right thing, we can have it all in one place associated with the chemical structure that you're synthesising.
9:53
With visualisations, we can kind of increase the complexity of this and create custom data comparisons.
9:57
So we can take say molecular properties, LogP, pKa, compare it with in vitro data within one view.
10:03
We have complex charts that we can build out here, a number of different options segmenting the legend by substructure, molecular weight.
10:10
We also have native filtering and outlier exclusion that you can use and then we can simplify lead optimization so we can create these comparison charts of the different molecules within your CDD Vault based on the properties.
10:23
So you can rapidly screen through what you're synthesising with these.
10:29
What's great about these visualisations is that they can be shared.
10:32
So you can save this search and then share it with your colleagues so that you can, you know, collaborate on this.
10:38
This has been useful for investor meetings or lab meetings so you can just run through the different save searches for the different projects you have going.
10:47
While I'm talking about sharing, I also want to mention the permission structure within CDD Vault is highly customizable.
10:52
So you'll have a vault admin that can control who sees what within each vault.
10:58
So users can be assigned to different projects.
11:01
And then you can also clone projects.
11:02
So molecules can be parts of different projects.
11:05
And say you have a CRO that you're working with, you can essentially duplicate a project.
11:09
So they have access to just that project.
11:11
They can upload the data.
11:12
It then gets automatically populated through all of your saved searches.
11:16
So you can see that, but they can't see any data that you don't want them to.
11:19
So it's a really great single source of truth for all the collaborations.
11:23
And that goes, if you've got collaborators at other institutions, we can build our projects and substructure that.
11:31
So with the activity and visualisation, we're going to create automatic insights and kind of see what's happening at the project level.
11:39
So how do we kind of create a whole workflow?
11:41
If you're using tools outside of CDD Vaults or you want to access that information, we have a Restful API that can essentially connect everything together and fill in the gap for your whole workflow.
11:52
So we can essentially take the data straight from the instrument, map it into the CDD Vault environment, and then make it available to your other tools.
12:01
So we can really fill in the gaps.
12:03
So everyone talks about AI, of course.
12:06
So the way we've approached AI is to enhance your existing workflows.
12:12
We're not creating anything entirely new.
12:14
We're using tools that have existed and are very stable.
12:18
So we can basically take your input molecule and through a number of steps, break it down into the latent vector that describes that.
12:25
And then we're going to take your latent vector and cross reference it to the databases you're familiar with.
12:30
So we've got 10s of millions of compounds here to do the similarity search.
12:34
And then we're going to, through trickery that the dev team has built out, this happens very, very quickly.
12:39
So within a few milliseconds, we're going to tell you that 467 hits are very similar to the latent vector that we input.
12:49
And where you do have a similarity score that I'm not showing there.
12:53
We can refine that down so if you want to look at compounds that have been previously patented, we can do that.
12:57
So just streamline your workflow here.
13:00
And then the main thing I want to add here is that your structure never left the CDD Vault environment.
13:08
All those databases of latent vectors came to your secure environment, so there isn't a shadow of your proprietary structure sitting on some database that you don't want it to be in.
13:20
We can take this another layer further with our bioisosteric suggestion, so we can fragment your molecule into its subcomponents.
13:28
You can pick which one you want to generate fragments for and we're going to show you of results based on which one you select.
13:35
So you can customise your predicted properties and then we're going to prioritise the synthesisable results coming out of this search.
13:42
So you're, we're not going to provide you, you know, 1000 compounds that you'd never be able to synthesise these are all fragments that have been created in the past.
13:49
We have also brought in other tools into the CDD Vault environment.
13:55
So with Alpha Full 2, you can generate PDB images and folding scores within CDD Vaults.
14:01
This image isn't going to move, but this is fully interactive.
14:05
You can play with it, move it after you've generated it, zoom in and export exactly what view you want.
14:10
You could then take that PDB image from Alpha Full 2 loaded into DiffDock and then dock molecules within the CDD Vault using DiffDock to get your docked ligand and docking score and then associate that with the chemical structures.
14:28
Brand new, just released this three weeks ago.
14:31
We brought in Boltz-2, so you can upload your FAST to your CIF and then you can train protein folding directly on the data within your CDD Vault environment and get those rendering scores available.
14:43
So how does this work as a whole enhanced workflow?
14:47
So we could take your small molecule library, we can then do bioisosteric suggestion on that enhance, do the enhanced search to get previously patented compounds, cross reference that.
14:58
So now we've got a new bank of compounds that have not been patented.
15:03
We can then go take your target of interest, run through AlphaFold 2 to get that PDB image, and then we can run both those through DiffDock or Boltz-2 to get those bolt docking scores.
15:14
And then we have a prioritised list for you to send to your CRO to go get synthesised.
15:18
They can then put that data directly into CDD Vault and enhance that workflow.
15:22
So the main thing to note here is that these are all turnkey AI tools.
15:26
You didn't have to bring someone in that knows how to do modelling.
15:29
These are already ready to go.
15:31
Your current team can deploy these, so it's really easy to use.
15:35
So at the end of the day, why would you want to work with CDD Vault?
15:40
And it's because of these folks.
15:41
This is our A-Team, our support team.
15:44
So they're the ones that are going to get you set up with your CDD Vault environment and support you whenever you have questions.
15:50
I want to mention that we offer a free demo, and the level of support you get during your demo is exactly the same as you get if you're paying for the software.
15:59
We really want to enable you and empower you to use this and get the most out of it and make sure that it's going to be a resource that's useful for you.
16:05
And this team is the reason that quarter over quarter we have an incredibly high customer satisfaction score.
16:12
People don't really leave CDD Vaults unless they get acquired or the company runs out of money.
16:19
So please come talk to us and we'd love to get you set up with CDD Vault.
16:24
The way to start or take the next steps is to start this conversation.
16:28
We've got Janice was floating around today.
16:31
Keith and then Bill are at booth 27.
16:34
And you can also go ahead and scan this QR code, and it will take you to the website so you can learn more and schedule a demo with us.
16:42
And with that I can take any questions if anyone has any.
