Extended Data Fig. 1: FIND-seq workflow details and sorted transcriptome purity.

(a–d) Transcriptome recovery and HIV gag DNA detection steps in FIND-seq, including (a,b) capture of single-cell genomes and transcriptomes in agarose, (c) reverse-transcription of polyadenylated RNA in agarose following oil removal and washes, generating single-cell agarose hydrogel beads with retained genomic DNA and covalently linked whole transcriptome cDNA that incorporates a template-switch oligonucleotide (TSO) for subsequent WTA, and (d) HIV detection PCR using gag primers and hydrolysis probe after hydrogel re-encapsulation on Device 2 (Fig. 1). (e) Diagram of whole transcriptome amplification and library preparation from material sorted by FIND-seq, with representative microelectrophoresis traces. (f,g) Purity of transcriptome material sorted by FIND-seq. (f) HIV DNA⁺ J-Lat human T cells and HIV DNA⁻ 3T3 mouse fibroblasts were mixed and subjected to FIND-seq followed by whole transcriptome sequencing. (g) Percentages of transcriptome reads from pure input cell populations, the 1:1 mixture of 3T3 and J-Lat, and HIV DNA⁺ cell transcriptomes sorted by FIND-seq that aligned unambiguously to human (red) or mouse (black) references. Results from a single experiment are shown.