Lightnews — Scholar-powered news

@georghochberg.bsky.social

200 followers 76 following 48 posts

Evolutionary biochemist and protein complex enthusiast at the University of Marburg.

Posts Media Videos Starter Packs

Pinned

georghochberg.bsky.social @georghochberg.bsky.social · Jun 11

New paper from the lab from Sriram Garg in my group. We introduce a general substitution matrix for structural phylogenetics. I think this is a big deal, so read on below if you think deep history is important. academic.oup.com/mbe/advance-...

A general substitution matrix for structural phylogenetics.

Abstract. Sequence-based maximum likelihood (ML) phylogenetics is a widely used method for inferring evolutionary relationships, which has illuminated the

academic.oup.com

3 52 95

georghochberg.bsky.social @georghochberg.bsky.social · 13d

A new preprint from the lab. I won't do a tweetorial until this is peer reviewed, but I think it's a banger, led by
@JedNzy. It's about Rubisco and what chaperones are really for. He's on the market, get him while you can. www.biorxiv.org/content/10.1...

Origin of chaperone dependence and assembly complexity in Rubisco biogenesis

Molecular chaperones assist with the folding and assembly of protein clients. Consequently, they are essential to diverse cellular functions. In most aerobic photosynthetic organisms such as B-cyanoba...

www.biorxiv.org

14 28

georghochberg.bsky.social @georghochberg.bsky.social · Jul 1

The Beeby lab's magnum opus on the Campylobacter flagellum is out. This is a uniquely powerful flagellum and how it evolved out of simpler precursors will keep us busy for a while still. Check it out! www.nature.com/articles/s41...

In situ structure of a bacterial flagellar motor at subnanometre resolution reveals adaptations for increased torque - Nature Microbiology

In situ cryoEM and modelling of the Campylobacter jejuni flagellar motor characterize the periplasmic scaffold and the structural basis for increased torque generation.

www.nature.com

1 8

georghochberg.bsky.social @georghochberg.bsky.social · Jun 26

Franziska did a spectacularly successful PhD in my group and to a large extend set up the lab when we first started. I'm extremely happy for her (though not the least bit surprised) that she won these very important awards for her PhD. Watch this one, great things are certain to come.

Max Planck Institute for Terrestrial Microbiology @mpimicrobiomarburg.bsky.social · Jun 26

📢 Congratulations 𝗗𝗿. 𝗙𝗿𝗮𝗻𝘇𝗶𝘀𝗸𝗮 𝗦𝗲𝗻𝗱𝗸𝗲𝗿 for receiving the Otto Hahn Medal and Otto Hahn Award from @maxplanck.de ! 🎉 The honors recognize her exceptional work @georghochberg.bsky.social. 🌟 Exciting times ahead! #MaxPlanck #ResearchExcellence www.mpi-marburg.mpg.de/1511259/2025...

Triple Honours for Franziska Sendker

Dr Franziska Sendker, a former doctoral candidate at the Max Planck Institute for Terrestrial Microbiology, has been awarded the Otto Hahn Medal by the Max Planck Society. The medal honours the outstanding achievements of young scientists and comes with a prize of 7,500 euros. In March 2025, she received the Bayer Pharmaceuticals Doctoral Award from the Society for Biochemistry and Molecular Biology (GBM e.V.), presented at the Mosbach Colloquium. Franziska Sendker's research showed that complex protein forms can arise not only through natural selection, but also through random genetic changes.

www.mpi-marburg.mpg.de

1 5

georghochberg.bsky.social @georghochberg.bsky.social · Jun 12

That is an alignment scoring matrix, similar to Blosum for amino acids. Alignment and phylogenetic substitution matrices have different purposes (one scores co-occurences within a column, the other the likelihood of substitution across a branch).

1 1

georghochberg.bsky.social @georghochberg.bsky.social · Jun 11

This matrix comes with the same simplifying assumptions we make about amino acid evolution, so there is that.
It will probably work for ABC transporters, but shoot me an email if you need help.

georghochberg.bsky.social @georghochberg.bsky.social · Jun 11

Exactly the opposite. This is an evolutionary model like WAG.

georghochberg.bsky.social @georghochberg.bsky.social · Jun 11

Oh and Sriram should be a co-corresponding author on this with me, but the journal messed this up after the proofs. I hope this will be corrected when the final version is out. He really drove this project.

georghochberg.bsky.social @georghochberg.bsky.social · Jun 11

Finally, I also want to point out the contribution of other groups in this field, in particular Nick Matzke and Christophe Dessimoz, who developed similar ideas to ours.

1 1

georghochberg.bsky.social @georghochberg.bsky.social · Jun 11

We are very excited about the possibilities this opens up, especially with respect to very deep history. We have number of very exciting things cooking that I’m dying to show to people when they are ready.

georghochberg.bsky.social @georghochberg.bsky.social · Jun 11

We are hoping that this matrix will eventually be integrated into IQtree directly, but for now you can just download it and use it as custom substitution model.

georghochberg.bsky.social @georghochberg.bsky.social · Jun 11

Our paper also discusses what we think this new method is good for, and what it may not be good for, as well as how it differs from other structural phylogenetics methods that other clever people have already developed.

georghochberg.bsky.social @georghochberg.bsky.social · Jun 11

We then use this new model to revisit the old question about the root using the structures of universal paralogs. Thankfully, the answer is that the root indeed lies where everyone always thought it would. But now we finally have certainty.

1 1

georghochberg.bsky.social @georghochberg.bsky.social · Jun 11

We inferred such a model from 3Di alignments of 1000s of protein families, so that you don't have to. The matrix is available with the paper, please feel free to use it immediately. You can use it easily with IQtree.

1 2

georghochberg.bsky.social @georghochberg.bsky.social · Jun 11

The key missing ingredient for this is a substitution model that describes how one 3Di character evolves into another. This is what our paper now provides.

1 3

georghochberg.bsky.social @georghochberg.bsky.social · Jun 11

This alphabet has 20 letters, and therefore integrates well with existing tree inference software (to which a 3di alignment looks like a regular old protein alignment). This means we can use this to infer trees.

1 1 4

georghochberg.bsky.social @georghochberg.bsky.social · Jun 11

The key development is the 3Di structural alphbet, developed by @martinsteinegger.bsky.social et al. 3Di allows a structural 'sequence' to be derived from a protein structure and was developed for remote homology searches.

1 4

georghochberg.bsky.social @georghochberg.bsky.social · Jun 11

The big problem was that we needed a model for the evolution of protein structures, such that we could unlock the power of maximum likelihood phylogenetics.

1 3

georghochberg.bsky.social @georghochberg.bsky.social · Jun 11

The idea is that instead of sequences, we could use structures to infer phylogenies. This idea is not ours, and it’s not new either. But for a long time there just wasn’t a good way to easily do this.

1 1

georghochberg.bsky.social @georghochberg.bsky.social · Jun 11

Here’s where our paper comes in. The central idea is that structure evolves more slowly than proteins. You can recognize structural homology over much greater phylogenetic distances than sequence homology.

georghochberg.bsky.social @georghochberg.bsky.social · Jun 11

This had long raised suspicions that the root may be wrong, or at least not certain, because our phylogenetic models struggle to find signal in these very diverged sequences. Other important problems, like the origin of photosynthesis suffer from the same issue.

1 1

georghochberg.bsky.social @georghochberg.bsky.social · Jun 11

There are a handful of such paralogs and that’s where the inference about the root of the tree of life came from. But it’s a messy business. The sequences of these universal paralogs have in most cases diverged almost beyond recognition.

georghochberg.bsky.social @georghochberg.bsky.social · Jun 11

The answer is so-called universal paralogs. These are genes that duplicated prior to LUCA, such that organisms now usually have two copies of them. If we throw both paralogs on the same tree, they root each other.

georghochberg.bsky.social @georghochberg.bsky.social · Jun 11

You might not know how this was actually decided. There is no outgroup for life, so how was that tree even rooted?

georghochberg.bsky.social @georghochberg.bsky.social · Jun 11

This is common and affects things you might think we know for sure. For example, I am sure you have seen the three (now two-) domain tree of life, that places the root (and therefore LUCA) on the branch between bacteria and Archaea.

georghochberg.bsky.social @georghochberg.bsky.social · Jun 11

For example, you might be interested in the relationship between two proteins that you suspect are related. But their lineages separated so long ago that their amino acid sequences no longer have any recognizable sequence similarity.