Split-GAL4 lines targeting PPL1

Asked at the VFB table, NeuroFly 2026 (Cologne), 8 September 2026.

The question

“Splits targeting PPL1?”

A typical table question: someone working on the dopaminergic PPL1 cluster wanted to know what split-GAL4 tools exist to target it.

How we answered it

1. search_terms for “PPL1” — resolves to the class dopaminergic PPL1 neuron (FBbt:00100219, symbol PPL1).
2. get_term_info + run_query (SplitsTargeting) on that class directly — this only found one split-GAL4 combination annotated at the whole-cluster level, which seemed too low given how well-studied PPL1 is.
3. run_query (SubclassesOf) on the same class — PPL1 has 18 named subtypes (the six individually numbered PPL101–PPL106, several SMP/dFB/MB-AMP/DP/MB-SV neurons, and the fan-shaped-body-projecting FB5H/FB6H/FB7B).
4. run_query (SplitsTargeting), batched across all 18 subtype IDs — this is where almost all of VFB’s split coverage actually lives.
5. For the four subtypes with zero splits, run_query (TransgeneExpressionHere) to check for non-split (single-transgene) driver lines instead, then run_query (FindStocks) on any transgene IDs found, to get orderable stock numbers.

What we found

Querying the whole class only surfaces splits explicitly annotated at that level — it does not aggregate splits recorded against its subtypes. That one whole-cluster result was P{ple-GAL4.DBD.TH-C} ∩ P{DAT-B-p65.AD}.

Splitting the query out by subtype tells a very different story:

Two of the zero-split subtypes (PPL102 and DP neuron) have no driver-line data of any kind in VFB — not even a single-transgene line, split or otherwise. For the other two, TransgeneExpressionHere did turn up single-transgene (non-split) lines, though with a caveat worth stating plainly: VFB’s own expressed_in annotation on those records points to other, more specific PPL1 subtypes rather than to MB-SV or the umbrella “mushroom body PPL1 neuron” grouping itself. Three of those single-transgene lines have orderable Bloomington stocks:

Worth being honest about: the high per-subtype counts (26–33) are not that many distinct tools — they’re mostly recombinations of a small, shared pool of hemidriver lines (DAT-B, ple-GAL4.DBD TH-C/D/F, R55C10, R60F07, R76F01, R76F02, R76F05, R61H03…) paired against each other, and a single split-GAL4 line commonly labels several PPL1 subtypes at once — so the per-row counts overlap heavily and don’t sum to a meaningful class-wide total. “SplitsTargeting” also only records an expression-pattern-to-class annotation, not whether a split is clean or specific to that one subtype.

Sources

• FBbt:00100219 — dopaminergic PPL1 neuron (PPL1)
• Xie et al., 2018, Cell Rep. 23(2):652–665; Aso et al., 2014, eLife 3:e04577/e04580; Aso and Rubin, 2016, eLife 5:e16135