Thank you, Klaus, and I'd like to thank Johannes and Alexander for all the work that's involved

in setting up a conference at such a nice location and beautiful place.

Thank you all for attending this last session and surviving the conference.

All right, let me get started.

Will this work?

Yes.

Okay, so I'm going to talk about some software developments that we've made in the Ultrascan,

SOMO, small angle, scattering modules.

We've been developing tools in the suite of programs for some years now.

The work on the size exclusion chromatography, SAX, online SAX setups and analysis involved

all three of these authors, Matthea, Javier Perez, Patrice Vachette, and myself.

But the new developments I'm talking about today are primarily Matthea's and mine.

A lot of the work has been motivated by Matthea and his desires to understand fibrinogen

and some of the reactions involved with that and its structure and how it works.

I've primarily contributed ideas to how these things should be done,

but I've been in charge of making it work in the software.

The software is available at somo.aucsolutions.com.

The US SOMO software is your main window,

and it has a lot of hydrodynamic calculations and capabilities.

Matthea talked about some of the advanced developments we're doing there.

A lot of it's centered around structures.

You load a PDB file, and there's some fibrinogen, and you'll get some...

Is there a laser pointer on here? How do I do that? Yay!

Then you get a collection of hydrodynamic parameters and structural parameters,

mainly structural at that point, when you load them up.

That's just to show you what data we're working with.

When you go to the SAS malls, SANS functions,

you get a lot more buttons to press.

If you work your way over to the HPLC kin,

which has been renamed because we have support for kinetics now.

Here's some fibrinogen data that was recorded. I assume it's Wing.

This is full fibrinogen, and it's an online SEC SACS setup.

This has been transformed already to an I of T collection.

Instead of looking at a collection of I of Q curves over time,

we're looking at a collection of I of T curves,

where each curve represents an individual Q value,

which makes it easier to decompose.

You have the obvious common centers of some sort of Gaussian

that can be used to fit at those points.

If you see you're not baseline separated,

then you probably do want to do some sort of decomposition

and not trust just pulling the peaks off

and assuming that is representative of your structure.

That's what our tools support.

You take a typical I of T curve at some Q value

and fit Gaussians to this.

There's not just regular symmetric Gaussians,

but there's standard skewed Gaussian support,

Gauss-modified Gaussian exponent.

There's multiple Gaussian methods.