Carlos Laborda

Designing proteins in public | Currently going after a hair-loss target 🥸

[Carlos Laborda @labordabio]

I'm going to try to design a new therapeutic protein for hair loss.

From my laptop.

No company. No lab of my own. No funding.

Just modern protein design tools, rented GPUs, published research and curiosity.

The idea is simple:

[Carlos Laborda @labordabio]

got distracted 🙂‍↔️

[Carlos Laborda @labordabio]

@jsaicecream Hi Jaber, sounds cool! Can you send me a message?

[Carlos Laborda @labordabio]

Day 5 of trying to design a therapeutic protein for hair loss.

What makes a protein surface attractive for binder design in the first place?

Finding hydrophobic patches is only part of the equation. If we want strong binders, we need to look beyond that:

[Carlos Laborda @labordabio]

1. a CRD / 73–86 hypothesis
2. another CRD surface hypothesis
3. an NTR / C-terminal hypothesis

Run them in parallel.

Then double down on whichever designs look most convincing.

Not one answer but small bets.

We’re still early. But that’s cool no?

[Carlos Laborda @labordabio]

Day 6 of designing a therapeutic protein for hair loss.

Ok cool. We need a surface that is:
> Accessible
> Structurally stable
> Continuous and large enough
> Hydrophobic but not too much
> And most importantly, biologically relevant.

Theory is nice but biology is much messier:

[Carlos Laborda @labordabio]

A mutation that weakens SFRP1’s Wnt antagonism doesn’t prove that region is where Wnt directly binds.

So, I don’t think the right move is committing to jus one epitope, throw a few hotspot residues into a model and pray.

At this stage the sensible approach is branching:

[Carlos Laborda @labordabio]

@Ruskinkot_chem delivery is def a big challenge. not just proteolysis but also how to make it through the skin barrier. A small molecule could be the right choice but first i'm trying to answer if neutralizing SFRP1 actually does what we want. Miniproteins are well stablished for that.

[Carlos Laborda @labordabio]

I'm going to try to design a new therapeutic protein for hair loss.

From my laptop.

No company. No lab of my own. No funding.

Just modern protein design tools, rented GPUs, published research and curiosity.

The idea is simple:

[Carlos Laborda @labordabio]

@menezespapi Glad you liked it Edu. I'll be rooting for you!

[Carlos Laborda @labordabio]

@LlorenteJandro Thanks Alejandro, we're just getting started! Claro, feel free 😊

[Carlos Laborda @labordabio]

@A_Cembellin_Pri That would be amazing Antonio! If we end up having something computationally convincing i'll take you up on that :)

[Carlos Laborda @labordabio]

Does binding there actually stop SFRP1 from doing its function?

Tomorrow.

[Carlos Laborda @labordabio]

> Chemically diverse. Hydrophobic residues can act as anchors, true, but surrounding polar and charged residues give you opportunities for hydrogen bonds, salt bridges and greater specificity.

And even if a surface checks every one of these boxes, there’s a harder question left:

[Carlos Laborda @labordabio]

Got rejected from Track 2 of the Anthropic x Adaptyv protein design competition.

I wonder how many independent researchers got in.

Anyways, bootstrapped it is then!

Makes for a good challenge: how to get designs into wet-lab testing without going bankrupt. 💰