Single-cell RNA-sequencing (scRNA-seq) has accelerated discovery in biology. Yet, current scRNA-seq methods miss potentially important transcriptional information due to limitations in molecular data analysis techniques. This thesis presents solutions to overcome some of these limitations. First, we developed TAR-scRNA-seq, a bioinformatics tool to quantify the expression of unannotated, cell-type specific transcripts. We applied TAR-scRNA-seq to the gray mouse lemur cell atlas and discovered over four thousand putative novel transcripts with cell-type specific expression. Second, we developed meta-scRNA-seq, a bioinformatics tool to uncover non-host transcriptomic expression in scRNA-seq data. We applied meta-scRNA-seq to over two million human and mouse cells to reveal sources of sample contamination and evidence of cell-specific infection. Third, we developed DART-seq, a molecular technique to enable simultaneous multiplexed amplicon sequencing and transcriptome profiling in single cells. We used DART-seq to capture the immune repertoire of human B cells with improved efficiency compared to conventional methods. Altogether, this thesis presents single-cell transcriptomics-based methods that enable the quantification of non-standard and unexpected RNA species of biological significance.