Skip to main content
. 2021 Feb 24;10:e60321. doi: 10.7554/eLife.60321

Figure 3. CRF_ID framework predicts identities for gene expression pattern analyses.

Figure 3.

(A) (Top) Schematic showing a fluorescent reporter strain with GFP expressed in cells for which names need to be determined. Since no candidate labels are known a priori neurons labels are predicted for all cells marked with pan-neuronally expressed RFP using full whole-brain atlas. (Bottom) A proxy strain AML5 [rab-3p(prom1)::2xNLS::TagRFP; odr-2b::GFP] with pan-neuronal RFP and 19 cells labeled with GFP was used to assess prediction accuracy. (B) CRF_ID framework with relative position features outperforms registration method (n = 21 animals) (***p<0.001, Bonferroni paired comparison test). Accuracy shown for top five labels predicted by both methods. Experimental data comes from strain AML5. Top, middle, and bottom lines in box plot indicate 75th percentile, median, and 25th percentile of data, respectively.