The AlphaFold2 Method Paper: A Fount of Good Ideas

Just over a week ago the long-awaited AlphaFold2 (AF2) method paper and associated code finally came out, putting to rest questions that I and many others raised about public disclosure of AF2. Already, the code is being pushed in all sorts of interesting ways, and three days ago the companion paper and database were published, where AF2 was applied to the human proteome and 20 other model organisms. All in all I am very happy with how DeepMind handled this. I reviewed the papers and had some chance to mull over the AF2 model architecture during the past couple of months (it was humorous to see people suggest that the open sourcing of AF2 was in response to RoseTTAFold—it was in fact DeepMind’s plan well before RoseTTAFold was preprinted.) In this post I will summarize my main takeaways about what makes AF2 interesting or surprising. This post is not a high-level summary of AF2—for that I suggest reading the main text of the paper, which is a well-written high-level summary, or this blog post by Carlos Outeiral. In fact, I suggest that you read the paper, including the supplementary information (SI), before reading this post, as I am going to assume familiarity with the model. My focus here is really on technical aspects of the architecture, with an eye toward generalizable lessons that can be applied to other molecular problems.

AlphaFold2 @ CASP14: “It feels like one’s child has left home.”

The past week was a momentous occasion for protein structure prediction, structural biology at large, and in due time, may prove to be so for the whole of life sciences. CASP14, the conference for the biennial competition for the prediction of protein structure from sequence, took place virtually over multiple remote working platforms. DeepMind, Google’s premier AI research group, entered the competition as they did the previous time, when they upended expectations of what an industrial research lab can do. The outcome this time was very, very different however. At CASP13 DeepMind made an impressive showing with AlphaFold but was ultimately within the bounds of the usual expectations of academic progress, albeit at an accelerated rate. At CASP14 DeepMind produced an advance so thorough it compelled CASP organizers to declare the protein structure prediction problem for single protein chains to be solved. In my read of most CASP14 attendees (virtual as it was), I sense that this was the conclusion of the majority. It certainly is my conclusion as well.

The Future of Protein Science will not be Supervised

But it may well be semi-supervised.

For some time now I have thought that building a latent representation of protein sequence space is a really good idea, both because we have far more sequences than any form of labelled data, and because, once built, such a representation can inform a broad range of downstream tasks. This is why I jumped at the opportunity last year when Surge Biswas, from the Church Lab, approached me about collaborating on exactly such a project. Last week we posted a preprint on bioRxiv describing this effort. It was led by Ethan Alley, Grigory Khimulya, and Surge. All I did was to enthusiastically cheer them on, and so the bulk of the credit goes to them and George Church for his mentorship.

AlphaFold @ CASP13: “What just happened?”

Update: An updated version of this blogpost was published as a (peer-reviewed) Letter to the Editor at Bioinformatics, sans the “sociology” commentary.

I just came back from CASP13, the biennial assessment of protein structure prediction methods (I previously blogged about CASP10.) I participated in a panel on deep learning methods in protein structure prediction, as well as a predictor (more on that later.) If you keep tabs on science news, you may have heard that DeepMind’s debut went rather well. So well in fact that not only did they take first place, but put a comfortable distance between them and the second place predictor (the Zhang group) in the free modeling (FM) category, which focuses on modeling novel protein folds. Is the news real or overhyped? What is AlphaFold’s key methodological advance, and does it represent a fundamentally new approach? Is DeepMind forthcoming in sharing the details? And what was the community’s reaction? I will summarize my thoughts on these questions and more below. At the end I will also briefly discuss how RGNs, my end-to-end differentiable model for structure prediction, did on CASP13.

Protein Linguistics

For over a decade now I have been working, essentially off the grid, on protein folding. I started thinking about the problem during my undergraduate years and actively working on it from the very beginning of grad school. For about four years, during the late 2000s, I pursued a radically different approach (to what was current then and now) based on ideas from Bayesian nonparametrics. Despite spending a significant fraction of my Ph.D. time on the problem, I made no publishable progress, and ultimately abandoned the approach. When deep learning began to make noise in the machine learning community around 2010, I started thinking about reformulating the core hypothesis underlying my Bayesian nonparametrics approach in a manner that can be cast as end-to-end differentiable, to utilize the emerging machinery of deep learning. Today I am finally ready to start talking about this long journey, beginning with a preprint that went live on bioRxiv yesterday.

The State of Probabilistic Programming

For two weeks last July, I cocooned myself in a hotel in Portland, OR, living and breathing probabilistic programming as a “student” in the probabilistic programming summer school run by DARPA. The school is part of the broader DARPA program on Probabilistic Programming for Advanced Machine Learning (PPAML), which has resulted in a great infusion of energy (and funding) into the probabilistic programming space. Last year was the inaugural one for the summer school, one that is meant to introduce and disseminate the languages and tools being developed to the broader scientific and technology communities. The school was graciously hosted by Galois Inc., which did a terrific job of organizing the event. Thankfully, they’re hosting the summer school again this year (there’s still time to apply!), which made me think that now is a good time to reflect on last year’s program and provide a snapshot of the state of the field. I will also take some liberty in prognosticating on the future of this space. Note that I am by no means a probabilistic programming expert, merely a curious outsider with a problem or two to solve.

What Does a Neural Network Actually Do?

There has been a lot of renewed interest lately in neural networks (NNs) due to their popularity as a model for deep learning architectures (there are non-NN based deep learning approaches based on sum-products networks and support vector machines with deep kernels, among others). Perhaps due to their loose analogy with biological brains, the behavior of neural networks has acquired an almost mystical status. This is compounded by the fact that theoretical analysis of multilayer perceptrons (one of the most common architectures) remains very limited, although the situation is gradually improving. To gain an intuitive understanding of what a learning algorithm does, I usually like to think about its representational power, as this provides insight into what can, if not necessarily what does, happen inside the algorithm to solve a given problem. I will do this here for the case of multilayer perceptrons. By the end of this informal discussion I hope to provide an intuitive picture of the surprisingly simple representations that NNs encode.

It is tempting to assume that with the appropriate choice of weights for the edges connecting the second and third layers of the NN discussed in this post, it would be possible to create classifiers that output over any composite region defined by unions and intersections of the 7 regions shown below.

∞

ICSB 2013

I recently had the pleasure of attending the 14^th International Conference on Systems Biology in Copenhagen. It was a five-day, multi-track bonanza, a strong sign of the field’s continued vibrancy. The keynotes were generally excellent, and while I cannot help but feel a little dismayed by the incrementalism that is inherent to scientific research and that is on display in conferences, the forest view was encouraging and hopeful. This is one of the most exciting fields of science today.

10 Months at Harvard, Quantified

I will soon reach the one-year mark of my fellowship at HMS, which seems like a fitting time to examine how effectively I have spent my time here so far. I have been a practitioner of self quantification long before the movement acquired its name, having tracked some aspect of my life since I was 16. Given the movement’s growing popularity, I thought it appropriate to share some of my life hacking experiments. My approach has cyclically peaked and waned in sophistication, something that I will expound upon later in the post, but I believe that the overall trajectory of my effort has been that of increasing usefulness. Any lifestyle change, particularly one that involves compulsive tracking of one’s behavior, ought to result in actionable information that is demonstrably useful and not merely be a quantitative exercise in vanity. In this post I hope to show that this can in fact be the case for self quantification.