Acknowledgements

Extended data

Author contributions

C.B.G.-B., S.D.W. and S. Aerts conceived the study. C.B.G.-B. developed pycisTopic, C.B.G.-B. and S.D.W. co-developed pycisTarget and the SCENIC+ modules and workflow and G.H. developed the code to generate custom cisTarget databases. C.B.G.-B. and G.H. made the SCENIC+ motif collection. C.B.G.-B. and S.D.W. performed the computational analyses, with the assistance of G.H., N.H. and S. Aibar. I.M. and S.P. generated the mouse cortex multiome data and J.W. performed the single-cell ATAC-seq experiments on the melanoma cell lines, with the assistance of V.C. C.B.G.-B., S.D.W. and S. Aerts wrote the manuscript.

Peer review

Data availability

Data generated in this manuscript, namely scATAC-seq in melanoma cell lines, 10x multiome in the mouse cortex and scATAC-seq in the Drosophila eye disc, are available in GEO under accession code {"type":"entrez-geo","attrs":{"text":"GSE210749","term_id":"210749"}} GSE210749 . GRCh38.86 genome annotation used in this study is available at https://ftp.ensembl.org/pub/release-86/gtf/homo_sapiens/Homo_sapiens.GRCh38.86.chr.gtf.gz. The GRCh38 genome index used in this study is available at https://cf.10xgenomics.com/supp/cell-arc/refdata-cellranger-arc-GRCh38-2020-A-2.0.0.tar.gz . The mm10 genome index used in this study is available at https://cf.10xgenomics.com/supp/cell-arc/refdata-cellranger-arc-mm10-2020-A-2.0.0.tar.gz . Data from ENCODE deeply profiled cell lines were downloaded from https://www.encodeproject.org/ , including bulk RNA-seq and ATAC-seq for eight cell lines, namely MCF7 ( ENCFF136ANW and ENCFF772EFK , for RNA-seq and ATAC-seq, respectively), HepG2 ( ENCFF660EXG and ENCFF239RGZ ), PC3 ( ENCFF874CFD and ENCFF516GDK ), GM12878 ( ENCFF626GVO and ENCFF415FEC ), K562 ( ENCFF833WFD and ENCFF512VEZ ), Panc1 ( ENCFF602HCV and ENCFF836WDC ), IMR90 ( ENCFF027FUC and ENCFF848XMR ) and HCT116 ( ENCFF766TYC and ENCFF724QHH ); and Hi-C data on five of the cell lines (IMR90 ( ENCFF685BLG ), GM12878 ( ENCFF053VBX ), HCT116 ( ENCFF750AOC ), HepG2 ( ENCFF020DPP ) and K562 ( ENCFF080DPJ )). STARR-seq data were downloaded from ENCODE ( ENCFF045TVA (K562), ENCFF047LDJ (HepG2), ENCFF428KHI (HCT116), ENCFF826BPU (MCF7)). ChIP-seq bigWig and summit bed files were downloaded from ENCODE using the following accession numbers: ENCFF702MTT and ENCSR000BHD for PAX5; ENCFF107LDM and ENCSR000BGU for EBF1; ENCFF803HIP and ENCFF934JFA for POU2F2 for bigWig and summit bed files respectively. The bulk RNA-seq experiments upon perturbation in these cell lines and ChIP-seq datasets are described in Supplementary Tables 1 and 4 , respectively. The 10x multiome data on PBMCs were downloaded from the 10x website. scRNA-seq data of baseline MM-lines and bulk RNA-seq data after SOX10 knockdown were downloaded from GEO ( {"type":"entrez-geo","attrs":{"text":"GSE134432","term_id":"134432"}} GSE134432 ). MITF, SOX10 and TFAP2A ChIP-seq data were downloaded from GEO ( {"type":"entrez-geo","attrs":{"text":"GSE61965","term_id":"61965"}} GSE61965 (MITF and SOX10) and {"type":"entrez-geo","attrs":{"text":"GSE67555","term_id":"67555"}} GSE67555 (TFAP2A)). SNARE-seq2 data on the human cortex were downloaded from Bakken et al. ⁶⁰ scATAC-seq and scRNA-seq data from the Drosophila eye-antennal disc were downloaded from GEO ( {"type":"entrez-geo","attrs":{"text":"GSE115476","term_id":"115476"}} GSE115476 ). The 10x Visium data and 10x single-cell multiome data from the human cerebellum were downloaded from the 10x website. All analyses can be explored in SCope ( https://scope.aertslab.org/#/scenic-v2 ) and UCSC in the following sessions: PBMCs ( https://genome-euro.ucsc.edu/s/Seppe%20De%20Winter/scenicplus_pbmc ), ENCODE cell lines ( https://genome.ucsc.edu/s/cbravo/SCENIC%2B_DPCL ), melanoma ( http://genome-euro.ucsc.edu/s/Seppe%20De%20Winter/scenicplus_mix_melanoma ), mouse and human cortex ( https://genome-euro.ucsc.edu/s/cbravo/SCENIC%2B_Cortex ), eye-antennal disc ( http://genome.ucsc.edu/s/cbravo/SCENIC%2B_EAD ) and human cerebellum ( https://genome-euro.ucsc.edu/s/cbravo/SCENIC%2B_cerebellum ). The SCENIC+ motif collection is available at https://resources.aertslab.org/cistarget/motif_collections .

Code availability

pycisTopic is available at https://github.com/aertslab/pycisTopic and deposited in Zenodo at 10.5281/zenodo.7857024. pycisTarget is available at https://github.com/aertslab/pycistarget and deposited in Zenodo at 10.5281/zenodo.7857022. SCENIC+ is available at https://github.com/aertslab/scenicplus and deposited in Zenodo at 10.5281/zenodo.7857017. Detailed tutorials and documentation on the SCENIC+ workflow are available at scenicplus.readthedocs.io and tutorials on pycisTopic and pycisTarget (within the SCENIC+ workflow and as standalone packages) are available at pycisTopic.readthedocs.io and pycistarget.readthedocs.io , respectively. Code to generate custom cisTarget databases is available at https://github.com/aertslab/create_cisTarget_databases . Our implementation of Cluster-Buster is available at https://github.com/ghuls/cluster-buster/tree/change_f4_output . Notebooks to reproduce the analyses presented in this manuscript are available at https://github.com/aertslab/scenicplus_analyses .

Competing interests

The authors declare no competing interests.

Footnotes

References