← Back to Passion Projects Completed

STICR Project: Do Sibling Neurons Share a Fate?

When a progenitor cell in the developing brain divides, do its descendants all become the same type of neuron, or do they spread out into different ones? STICR is a lineage-tracing method that answers this by tagging a cell with a DNA barcode that every one of its descendants inherits. Cells sharing a barcode form a clone, so cell type and family history can be read out together.

This project analyzed the published STICR data from the ground up: choosing which of the 13 available libraries were usable, clustering and annotating the cells, correcting for confounds, and finally testing how tightly lineage and fate are linked in cortical interneurons.

What I did

1. Picked the data. Each of the 13 libraries was scored on cell count, the share of cells linked to a clone, and clone size. Five libraries (CA238 to CA242) were clearly the strongest and were combined, with Harmony batch correction. The libraries blended rather than splitting into separate islands, which is the check that the correction worked.

2. Separated cell age from cell type. Clusters can reflect how mature a cell is as much as what it is. I built a maturity score from general immature-neuron genes, removed it, and re-clustered. A cluster first labeled as interneuron-lineage split in two: about 15% of its cells (472) turned out to be a striatal, non-interneuron population that had only been grouped with the interneurons because of developmental stage. The Sst+ and Vip+ populations stayed intact.

3. Zoomed in on interneuron candidates and removed cell-cycle effects. Narrowing to the interneuron-candidate clusters and correcting for cell-cycle stage produced 13 finer subclusters, including a confirmed Nkx2-1+ MGE-derived progenitor population and a confirmed Ndnf+ neurogliaform population. A second, gentler cell-cycle correction recovered both again, so the findings are not an artifact of one analysis choice. Both were also checked against spatial gene-expression images from the Allen Developing Mouse Brain Atlas.

4. Matched against a published reference. All 6,787 interneuron-candidate cells were mapped onto the Allen Institute's annotated mouse cortex dataset, narrowed to 4,469 reference interneurons. This gave official labels in place of marker-gene guesses, and recovered all seven subtypes plus each cell's developmental origin (2,403 MGE-derived, 4,374 CGE-derived).

Allen subtype	Cells matched	Share
Vip	2,867	42.2%
Sst	1,405	20.7%
Pvalb	1,147	16.9%
Lamp5	961	14.2%
Sncg	367	5.4%
Sst Chodl	24	0.4%
Meis2	16	0.2%

What I found

Sibling cells agree on their broad developmental origin far more often than on their exact subtype. Across 607 multi-cell clones, siblings shared the same fine subtype in 29.7% of clones, but stayed within the same broad origin (MGE versus CGE) in 57.8%. That pattern suggests fate is committed in stages: an early decision about which broad region a lineage belongs to, then a later, more flexible decision about the exact subtype.

A dataset detail worth fixing

The public record lists a single embryonic day 13 to 14 label for the whole dataset. Tracking down the source paper's per-library records showed that this is only when the lineage-tracing virus was injected. The tissue was collected after birth, between postnatal day 5 and day 15. That resolved an apparent contradiction in the data: Ndnf does not switch on until around embryonic day 18.5, so finding Ndnf+ cells is exactly what postnatal tissue should show.

Caveats

Cell type calls from the first pass were made by eye from marker genes and are a starting point, not a final answer. Reference matching confidence was moderate (median 0.57 on a 0 to 1 scale), expected when two datasets are collected with different methods. One clone pairing an MGE progenitor with a Vip+ cell, which is normally CGE-derived, still deserves a manual check.