2023 Retrospective
For me, 2023 was a year of a big change, moving to Australia and starting a new job. Scientifically, little has changed. I am dealing with the same questions as before: how to integrate large-scale microbiome data? how to deal with small protein sequences?1
Australia is treating me and my family very well: the weather is great, the food is good, the coffee is excellent. The Gods of Peer Review smiled on me and I got an ARC Future Fellowship to continue working on small proteins and an NHMRC grant to continue working on antibiotic resistance with most of the same fantastic group of people who were part of the EMBARK Programme.
We produced a few important papers in 2023: with César de la Fuente, we submitted the AMPSphere manuscript for publication, which was the first big output of the small proteins research project (building on the release of Macrel a few years ago). We will have another small proteins preprint within a week or so, describing how we’re making the full catalog of smORFs from the global microbiome (GMSC) available to the world.
SemiBin got an update, SemiBin2 which is not only much faster, but added support for long-reads. SemiBin is not a finished product, but continues to see ongoing improvements.
We have always committed to maintaining our tools for a long time and we discussed this in the Lab’s June 2022 newsletter. This led to a bit of an online discussion2 and I took on as a procrastination project3 to write a more formal perspective which also touched on how it’s actually implemented. This perspective was finished about a year later (August 2023) and is now being revised for resubmission (since it’s not original research, there is no preprint, but you can read it on Google Docs).4
With Shaojun Pan (first author of SemiBin), we have been working with Kristoffer Sahlin on adapting strobealign more suited for metagenomics. This all started when about a year ago, I read the Strobealign paper and it seemed suitable for something that I have been wanting to do for a long time: out of core mapping/abundance estimation. I emailed Kristoffer and we have been working together really well5! Personally, I have also really enjoyed going back to doing the type of algorithmic work I started my research career with. Back then, I decided to move into more applied problems to gain context, but it’s still very enjoyable to do some shape rotations.
We were also a part of some other great collaborations, which culminated in the SPIRE manuscript and a manuscript in Nature describing many novel gene families.
Looking forward to 2024, I will write more blogpost-like postings here on Substack. This is a personal project and it does not necessarily aim to advance my career.6
In parallel, I will also aim for more regular videos on YouTube as a professional goal: I think the videos have a good audience reach to effort ratio (even if we remain, on the global scheme of things, a very small channel), but they are rarely the priority in any particular week. So, they sit in the important/not-urgent quadrant for too long7.
Links of the Week
Age of Invention: Does History have a Replication Crisis? /ht @Sam_Dumitriu
A wonderful history of engines (with really nice animations)
Modern art is 90% very bad and outright tax fraud is rampant, but the old man shouting at clouds opinion that abstract art is all a fraud is wrong.
Photos of the Week
I’m recruiting PhD students interested in these topics!
Spread over at least 3 platforms, so I cannot find it anymore...
This is the thing you work on when you should be filling out some administrative forms.
In a recent Twitter Thread, I described how I completely failed to follow my own advice when it came to webservers!
At least I feel it so and I hope he feels the same!
I’m aiming for net neutral effect on career.
There is a world in which a research group spends too much time doing online videos/promotion and not enough time doing science, but me and my group are still in a position where the marginal return on videos is very high.