Specificity was the bottleneck. The design of the resource reflects that. Rather than hunting for one gene that marks one cell type, the Salk group screened several hundred engineered lines for expression patterns that partially overlapped, then used those intersections to build split-GAL4 combinations restricted to the overlap [7]. Asahina's framing is the honest version of why: "Virtually, no single gene is sufficient for specifying one single cell type" [6].
The combinatorics explain where the labor went. Take 300 lines as an illustrative reading of "several hundred": pairwise intersections number 300 x 299 / 2, or 44,850 [17]. Almost none of those pairs are useful, and no screen tells you which ones are without imaging each. The deliverable is the curated subset that survived the screen.
Two qualifiers in the coverage claim deserve to stay attached. The access is to nearly all *long-range* octopaminergic and tyraminergic neuron types [1], so cells that stay local are outside the stated scope, and "nearly all" is not a fraction anyone has to defend numerically. The thing this doesn't tell you is how much of the system's behavioral output rides on the types that did not make the cut.
The aggression result is the cleanest demonstration of why per-type access matters. Earlier work had tied octopamine signaling to both more and less aggression [9], which is the sort of literature that gets called a paradox when the unit of analysis is a molecule. With the drivers in hand, the team describes a previously uncharacterized type, ASM4, that suppresses aggression in socially isolated males and females [10]. That reframes the disagreement as heterogeneity among cells rather than inconsistency among experiments [11]. It does not close the chemical question: the authors allow that the divergence may also depend on whether a given cell signals through tyramine, octopamine, or both [11], and tyramine is octopamine's precursor, so the two are not independent knobs [3]. Whether ASM4 was silenced, activated, or both is exactly the kind of detail that sets how much weight the suppression claim can carry.
The visual work makes the same point without the historical baggage. Distinct optic-lobe-projecting types modulated optomotor stabilization and object tracking differently [12], which is Fajner's summary in one line: "one modulator, but distinct neurons, distinct jobs" [13]. Alongside that, the toolkit let the group register genetically defined types against electron-microscopy connectome data and compare male and female anatomy [8], which is the step that makes a driver line useful to someone who did not build it.
On humans, the authors are careful and so am I. Their bridge to vertebrate noradrenergic signaling rests on conserved neuromodulatory principles, not on any vertebrate measurement in this paper [14], and octopamine's status as a functional analog of noradrenaline is a description of its role in invertebrates [3]. A resource paper's value shows up in other people's experiments, on a timescale of years.
