Alignment results of all datasets (Table 1) run with either Cell Ranger or kallisto against Ensembl reference. a Percent alignment rates of reads against the reference transcriptome. b Total gene detection. c Median gene counts over all cells per dataset. d Median UMI counts over all cells per dataset. e Total cell counts of each dataset

Impact of reference choice and alignment pipeline on zebrafish scRNA-seq datasets. a Percent alignment rates of reads against the reference transcriptome. b Total gene detection. c Median gene counts over all cells. d Median UMI counts over all cells. e Total cell counts of each dataset

Violin-plots showing comparison of gene detection per cell between kallisto and Cell Ranger across selected examples (a, b, c, d, e, f). Cell Ranger datasets contain an additional cell population with lower gene detections, not present in the kallisto datasets (at and below dashed lines). This observation was not seen for the human datasets (f). The remaining datasets can be found in Additional file 1: Fig. S2 and S3

Downstream analysis of the dr_pineal_s2 dataset. a 2D visualization using UMAP of Cell Ranger analyzed clusters. Each point represents a single cell, colored according to cell type. The cells were clustered into 13 distinct types, which were defined according to their unique transcriptomes (Additional file 2). b Expression profile of marker genes according to cluster [7] of (a). Cone- and rod-like PhRs are defined according to their transducin protein subunits (gnat1 and gngt1 in rods vs. gnat2 and gngt2a in cones [26] and opn1lw1 and parietopsin in cone-like). c 2D visualization using UMAP of kallisto analyzed clusters. Each point represents a single cell, colored according to cell type. The cells were clustered into 14 distinct types, which were defined according to their unique transcriptomes (Additional file 2). d Expression profile of marker genes according to cluster [7] of (c). Two different populations of cone-like PhRs are detected. Both express the cone unique transducin protein-subunits, but differ in the expressed opsin (opn1lw1 in red-cones, and parietopsin in green-cones), demonstrating the bi-chromatic photoreception characteristic of the pineal gland. col14a1b was only detected in the kallisto dataset and is the strongest DE marker within the red-cone cluster (c, d)