MAHOMES can distinguish between metalloenzyme sites and metal binding sites

This work started with the problem that it’s hard to look at a cleft in a protein and predict if that cleft can do chemistry. We know there are millions of physical and chemical aspects of an enzyme that make catalysis happen. But which features could we use to predict catalysis? This is important for understanding enzymes. It’s also important for DESIGNING enzymes. Why? Well, it’s especially hard to predict if a designed enzyme cleft has activity or not because there are no markers of enzyme activity from evolution (because the amino acids in the cleft didn’t evolve they were designed).

We wanted to know if we could learn what the special sauce is that makes an enzyme active site catalytic.

Ryan Feehan and Meghan Franklin developed a machine learning model to tackle this challenge. When making our classifier the key decision was finding the non-enzymatic sites that we should train the model on. We decided to train the model to distinguish between metal binding sites and metal catalytic sites. That way the sites would be extremely similar other than that some could do catalysis and others couldn’t.


Making the dataset for this was a trial, but Ryan persevered on this task—taking it from an undergraduate research project through to a graduate research project! Ultimately, Ryan was able to find thousands of unique active and inactive metal binding sites. This dataset may be useful to others and so we made it available on github.

Fig. 1

The sites were from all kinds of enzymes, giving us hope that this may lead to generalizable enzyme features.

We tried all kinds of machine learning algorithms. Ryan named the top model Metal Activity Heuristic of Metalloprotein and Enzymatic Sites (MAHOMES). Incidentally, Ryan is a KC Chiefs fan. MAHOMES can distinguish between extremely similar metalloprotein sites and metalloenzyme sites with 92.2% precision and 90.1% recall using only physicochemical features!

We attribute this success to the training set we created which had two important characteristics:

1. The set was fairly balanced (more similar numbers of active and inactive examples)

2. The set used particular examples of active and inactive sites that were fairly similar to each other (all metal sites) so the model could pick out features that were truly important to catalysis

Overall, our physicochemical method compared well even in comparison to methods that used homology!

On a personal note, this is my first non membrane protein paper published in more than 20 years! I’m excited that the lab is going in new directions.

Membrane Barrels Are Taller, Fatter, Inside-Out Soluble Barrels

The title says it all, but let me elaborate just a bit. This paper came about because we often talk in lab about how outer membrane beta barrels are inside out soluble barrels– it’s a concept that’s intuitive but not obvious how to demonstrate. In contrast to membrane beta-barrels, membrane helices have long been known to NOT be inside out soluble proteins. Membrane helices have a hydrophobic core similar to soluble proteins, and a more hydrophobic exterior where they interact with the membrane. But membrane barrels have a hydrophilic core! Some of their cores are hydrophilic because the barrel is wide enough to be water solvated, and some of the barrels are small but have charged amino acids on the interior anyway. Rik Dhar compared the hydrophobicities of structurally characterized membrane barrels to similarly structured soluble barrels. He found that the hydrophobicity of the exterior of one perfectly matched the hydrophobicity of the interior of the other and visa versa

Another feature of barrels that is commonly mentioned but had never been proven is that the hydrophobicity of the amino acids alternate: polar, non-polar, non-polar, polar etc. I *think* the first mention of this is a 2002 review by Schulz, but there it’s mostly speculation.

The idea of alternation makes sense because the side chains in a beta strand alternate direction almost perfectly

But Ryan Feehan found that though the directional alternation is really high, the hydrophobicity alternation only occurs at ~74% efficiency.

Finally, this manuscript really took shape when Rik Dhar decided to use Covid time to learn to code! I think it’s amazing that he learned to code and published a paper using his new skills in less than a year. Way to go!

Folding TolC in vitro is concentration dependent

It’s tricky to refold membrane proteins! One particular challenge is to refold multi-chain β-barrels. So, often they have to be purified from native membranes. Our lab has a new paper out on folding the multi-chain β-barrel TolC which is the outer membrane part of an antibiotic resistance pump. Jimmy Budiardjo and Ayotunde Paul Ikujuni found that the trick to folding it was to concentrate it in the presence of a membrane mimetic. Undergraduate Andres Cordova saw that folding it this way in vitro maintained native function. Then Emre Firlar and Jason Kaelber showed that folding it this way in vitro had 2D class averages consistent with the native fold. The article is part of an Festschrift celebrating Steve White’s 80th birthday and his amazing contributions to understanding membrane protein folding.

Happy birthday, Steve!

Slusky Lab Statement

To the Black members of our community:

We want to acknowledge your pain and exhaustion during this difficult time.

To all members of our community:

We cannot be silent in the face of the trauma being inflicted on the Black community. We affirm that Black lives matter. We will not be silent about racism in our community from the police or from members within the scientific community.

With respect to police brutality:

•We want police accountability to be overseen by a third party.

•We want Kansas City to have local control of its own police department.

•We want to ensure the continued legality of the videotaping police and of making those videos public.

•We want a demilitarization of the police, including a complete stop to the use of teargas.

•We want taxpayer funds to be reallocated from police to social workers and mental health facilities.

•We demand that the police limit the use of excessive force, especially for unarmed people. We demand that the police partake in extensive training on de-escalation techniques and lethal force be used as a last resort or not at all.

We hope that soon science and the broader community will be equitable to Black, Indigenous, and People of Color and we commit to working toward that goal.

Joanna Slusky   Jimmy Budiardjo   Jaden Anderson    Gustavo Murillo-Espinoza

Daniel Montezano    Jakki Deay    Rebecca Bernstein   Paul Ikujuni   Ryan Feehan