Skip to content

EPIC: region-free ganglionic eminence (MGE/CGE/LGE) lineage classes for GABAergic neurons #3789

Open

Description

@cellsemantic

Summary

Ganglionic eminence (GE) lineage gives a general classification for telencephalic GABAergic neurons. CL's GE-lineage classes are currently either restricted to cerebral cortex or interneuron-only. Mouse and primate data both show GE-derived GABAergic types across cortical and subcortical regions:

  • the whole mouse brain taxonomy (Yao 2023);
  • the HMBA basal ganglia consensus taxonomy (CS20250428), whose classes are explicitly CN LGE GABA, CN MGE GABA and CN CGE GABA.

This epic asks for region-free lineage classes and the fixes needed to hang existing terms under them.

Evidence

Location. Fraction of each WMB subclass's MERFISH cells (Yao 2023) by division:

WMB subclass BG group mapped (MapMyCells) BG lineage Isocortex OLF HPF CTXsp STR PAL
046 Vip Gaba VIP GABA CGE .69 .09 .16 .05 .02
049 Lamp5 Gaba LAMP5-CXCL14 GABA CGE .70 .13 .12 .02 .03
053 Sst Gaba STR SST-ADARB2 GABA MGE .68 .07 .19 .06 .01
056 Sst Chodl Gaba STR SST-CHODL GABA MGE .20 .09 .02 .04 .49 .14
054 STR Prox1 Lhx6 Gaba STR FS PTHLH-PVALB GABA MGE .01 .07 .88 .03
055 STR Lhx8 Gaba STR TAC3-PLPP4 GABA MGE .80 .15
058 PAL-STR Gaba-Chol STR / STRd / GPin-BF Cholinergic GABA MGE .01 .44 .50
059 GPe-SI Sox6 Cyp26b1 Gaba GPe MEIS2-SOX6 GABA (arkypallidal) LGE .14 .82
061/062 STR D1/D2 Gaba STRd/STRv D1/D2 MSN LGE (striatal; MERFISH fractions not available for these nodes)

Provenance. The BG↔WMB correspondences come from a MapMyCells annotation transfer, summarised as a curated crosswalk; they will be re-checked against the raw MapMyCells output. Each BG lineage is the BG taxonomy's class assignment.

Consequence in current CL. The BG-derived primate terms in the striatum are inferred to be cerebral cortex neurons. For example, STR SST-CHODL GABA … (Primate) (CL:4310123), VIP GABA … (Primate) (CL:4310127) and LAMP5-CXCL14 GABA … (Primate) (CL:4310084) are all inferred subclasses of cerebral cortex neuron and cerebral cortex GABAergic interneuron, because their lineage/marker parents carry cortical soma location.

Requested changes

MGE

  • New: medial ganglionic eminence derived GABAergic neuron. GABAergic neuron ⊓ develops from some medial ganglionic eminence (UBERON:0004024), with no location. The genus is neuron, not interneuron, so it can cover MGE-derived projection and cholinergic types (e.g. striatal/pallidal cholinergic-GABAergic neurons).
  • Re-parent medial ganglionic eminence derived interneuron (CL:4023063) and medial ganglionic eminence derived GABAergic interneuron (CL:4023069, cortex) under it. CL:4023069 stays as the cortical subclass.

CGE

  • Logical definition for caudal ganglionic eminence derived neuron (CL:4042039), which currently has none: neuron ⊓ develops from some caudal ganglionic eminence (UBERON:0004026).
  • New: caudal ganglionic eminence derived GABAergic neuron, with no location.
  • Fix caudal ganglionic eminence derived interneuron (CL:4023064). It is asserted under cerebral cortex GABAergic interneuron, but CGE-derived VIP and LAMP5-CXCL14 types also occur in striatum (BG taxonomy). Either drop the cortical assertion, or rename/redefine it as the cortical subclass and add a region-free parent.

LGE

  • New: lateral ganglionic eminence derived GABAergic neuron, under existing lateral ganglionic eminence derived neuron (CL:4023035), which is already region-free. Medium spiny neurons and arkypallidal neurons would classify here.

Relationship to open requests

Not included (for discussion)

The BG taxonomy also has non-GE telencephalic/diencephalic GABAergic groupings, e.g. CN LHX8 GABA (prototypic GPe and basal-forebrain LHX6-LHX8-GBX1 types) and F M GATA3 GABA (nigral/VTR GATA3 projection types). Their developmental origin isn't stated, so marker-defined classes may be safer than lineage claims. They are left out of this epic.

Related: #3321 (Basal Ganglion product EPIC), #3566.

Activity

  1. cellsemantic commented on Oct 12, 2026

    @cellsemantic
    CollaboratorAuthor

    Related epics from the same analysis: #3788 (region + NT grouping classes), #3790 (lineage + marker classes, which depend on this one), #3791 (axiom errors).

  2. dosumis commented on Oct 13, 2026

    @dosumis
    Contributor

    @ai4cur-agent. Please implement all of these fixes in one PR.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions