PROJECT SUMMARY / ABSTRACT
Recent knowledge gained from studies based on scRNAseq provides new opportunities for the
characterization of cell types and description of phenotypes. In addition, information gained from scRNAseq
can help develop new reagents useful for functional studies. One of the main challenges in scRNAseq studies
to address is the mapping of scRNAseq clusters to anatomy. While this can be relatively easily done if a cluster
is defined by the unique expression of a single differentially expressed gene (DEG), the task is more
complicated if a group of cells can only be defined using a combination of DEGs. To identify combinations of
genes whose expression define cell clusters, we propose in Aim 1 to develop an algorithm that determines
the smallest unique set of marker genes that define a cluster. A number of approaches and resources can
be used to associate the expression of marker genes to the anatomy; however, in most cases co-expression of
multiple genes is required to properly map clusters to the anatomy, and existing reagents are limited. As the
split-Gal4 method, whereby the DNA-binding domain of Gal4 and a transcriptional activator domain are
expressed independently under the control of two different enhancers, is scalable for the detection of multiple
genes, in Aim 2, we propose to build on our established TRiP/DRSC platform and generate a collection
of 1012 split-Gal4 lines that will be used to map the estimated 450 cell clusters present in the fly.
Importantly, the split-Gal4 lines generated from these studies will allow us to address a long-standing issue in
the Drosophila field, namely the specificity of existing Gal4 lines, as we and others have observed that the
large majority of so-called “tissue-specific” Gal4 driver lines are also expressed in additional, unwanted cell
types or tissues. Finally, as the current split-Gal4 system is not compatible with temporal control, preventing
the use of split-Gal4 lines for experiments requiring stage specific activation of the UAS driven transgene, in
Aim 3, we will develop a robust split-Gal4 method that allows temporal control and generate a
collection of 200 inducible split-Gal4 lines that will allow the generation of 100 controllable tissue
specific Gal4 patterns, covering most larval and adult tissues. Together, these resources will provide the
community with much needed reagents that will be made available and distributed as soon as they are
produced by the Bloomington Drosophila Stock Center.