Author: Gregor J. Rothfuss

Designer Proteins

the baker lab / rosetta are still going strong, with complementary ML approaches:

The hope is that the next time there’s an outbreak, within 2 days, we’ll have models of candidates

2023-07-13: Protein design is getting real.

10–20% of RFdiffusion’s designs bind to their intended target strongly enough to be useful, compared with less than 1% for earlier, pre-AI methods.Biochemist Matthias Gloegl has been hitting success rates approaching 50%, which means it can take just a week or 2 to come up with working designs, as opposed to months. “It’s really insane”. The latest protein-design tools have proved to be extremely powerful at creating proteins that can do a particular task — so long as that function can be described in terms of a shape, such as the surface of a protein to bind to. Tools such as RFdiffusion aren’t yet able to handle other kinds of specifications, such as making a protein that can carry out a particular reaction regardless of its shape — when “you know what you want but you don’t know what the geometry is”.

Post-AI Chess

AI Ruined Chess. Now, It’s Making the Game Beautiful Again

Kramnik saw flashes of beauty in how AlphaZero adapted to the new rules. No-castling chess provoked rich new patterns for keeping the king safe. A more extreme change, self-capture chess, in which a player can take their own pieces, proved even more alluring. The rule effectively gives a player more opportunities to sacrifice a piece to get ahead, a tactic considered a hallmark of elegant play for centuries.

Blands

On all the garbage direct to consumer startups.

All startups seek to disrupt and disintermediate a smug status quo, or originate and dominate an entirely new niche. But what makes a brand a bland is duality: claiming simultaneously to be unique in product, groundbreaking in purpose, and singular in delivery, while slavishly obeying an identikit formula of business model, look and feel, and tone of voice.

Despite hiding in plain sight (and plain recycled packaging), this “slight of bland” has won the wallets of a generation that considers itself above marketing, and created some of the buzziest companies of the age.

2023-04-01: It’s worse, average is everywhere

The interiors of our homes, coffee shops and restaurants all look the same. The buildings where we live and work all look the same. The cars we drive, their colors and their logos all look the same. The way we look and the way we dress all looks the same. Our movies, books and video games all look the same. And the brands we buy, their adverts, identities and taglines all look the same.

But it doesn’t end there. In the age of average, homogeneity can be found in an almost indefinite number of domains.

The Instagram pictures we post, the tweets we read, the TV we watch, the app icons we click, the skylines we see, the websites we visit and the illustrations which adorn them all look the same. The list goes on, and on, and on.

There are many reasons why this might have happened.

Perhaps when times are turbulent, people seek the safety of the familiar. Perhaps it’s our obsession with quantification and optimization. Or maybe it’s the inevitable result of inspiration becoming globalized.

Regardless of the reasons, it seems that just as Komar and Melamid produced the “people’s choice” in art, contemporary companies produce the people’s choice in almost every category of creativity.

Unintuitive RNA

RNA are much larger than Proteins

But we don’t have to get into such details in order to see how “off” our perceptions are. I, and probably many others, have a mental picture of an mRNA sequence being read out and transcribed into a larger protein, but that can’t be right. And it isn’t! In reality, the mRNAs are ~10x the size of proteins themselves. This feels wrong to us, because we’re used to our own preferences for algorithmic compression: blueprints are smaller than houses, but not inside the cell.