I studied under a professor who worked with Woese. It was great- lots of emphasis on RNA, using RNA to answer basic biological questions. One of the things I've never been able to understand is, how did Woese, with just a limited number of very limited sequence data, manage to carve out the existence of an entirely new form of life? I ask, because looking at all the effort to sequential gigabases today, it doesn't seem like we're learning geometrically more knowledge. Modern gains in sequencing seem highly incremental, and generally only provide correlation data, rather than causal data, and the data they do provide is not actionable.
Was Woese's discovery just "easy"? Would any competent scientist working in the field make that? Where are the Woeses of sequencing today?
In a way, yes. It’s widely established that 20th century biologists got to pick the low-hanging fruits. This shouldn’t detract from the sheer genius they displayed.
In the specific case of the three domains of life (bacteria, archaea and eukaryotes) — which I’m assuming you’re referring to? — Woese & al. got “lucky” with their study of 16S rRNAs.
In a nutshell, these RNAs are relatively easy to sequence and analyse. Their extremely high degree of conservation means that they can be trivially aligned to each other to study differences. But they also contain “hypervariable” regions that made it possible to distinguish different species at extremely high resolution, even using relatively slow computers and unsophisticated analytical tools. In addition, modern phylogeny mostly struggles with inconsistencies in classical methods that are due to unaccounted features (such as lateral gene transfer, which invalidates the typical assumption of a strict tree structure). Early studies ignored these inconsistencies (or didn’t even notice them, due to the scarcity of data).
But since you talk about correlative vs causative data, it’s important to note that this fundamentally hasn’t changed: from a very high-level perspective, new evidence is either compatible with existing theories, or isn’t. Strictly speaking, evidence never demonstrates causation — it can’t. But sometimes there are few viable alternative models, so the available evidence trivially allows discarding all but one or two models. This gets harder the more nuanced your models are, and this is the natural progression we have in any scientific endeavour.
Based on my view of Woese's work (in which I saw the years of effort leading to better secondary structure alignments), I don't think it was "easy" nor was it a low hanging fruit.
And I think there's still low hanging fruit that people are missing.
… compared to subsequent phylogenetics (we’re all standing on the shoulders of giants, after all). I agree that, at the time, the work must have been (a) mind-numbingly tedious, (b) intellectually challenging.
But comparing past papers (including that one [1]) to current ones does show that the amount of data, computation, and evidence necessary to produce a contemporary research paper is incomparably larger than it used to be. The seminal DNA structure paper was a single A4 page, with an estimate of the double helix height, no calculations or data shown. Good luck getting something like this published today.
Was Woese's discovery just "easy"? Would any competent scientist working in the field make that? Where are the Woeses of sequencing today?