Overview

This directory contains source code for reproducing the analysis in:

Sutton, G.J., Poppe D., Simmons R.K., Walsh K., Nawaz U., Lister R., Gagnon-Bartsch J.A., and Voineagu I. Comprehensive evaluation of deconvolution methods for human brain gene expression. Nat Commun 13, 1358 (2022). https://doi.org/10.1038/s41467-022-28655-4

File naming conventions are as follows, and should be followed in order:

• Preprocessing: scripts that preprocess public and generated data
• BenchmarkingDatasets: estimating composition in simulated mixtures, and analyses of its output. Broadly corresponds to Figures 1-3. RNA: In vitro RNA mixtures of cultured neuronal and astrocyte RNA. DM: Mixtures derived by simulating pseudobulks from Darmanis et al.'s single cell RNA-seq. VL and CA: Mixtures derived by simulating pseudobulks from Velmeshev et al.'s or the Human Cell Atlas' single nucleus RNA-seq (VL and CA, respectively). Zhang: deconvolution of Zhang et al.'s immunopanned pure brain cell-types.
• ConfoundingComposition: simulation of datasets with confounded composition between groups, exploring DE driven by this phenomenon. Broadly corresponds to Figures 4-5. CompositionOnly: differential expression driven purely by group confounds in composition. CompositionAndExpression: detection of differential expression when confounded by composition. CIBx: analyses using CIBERSORTx to impute cell-type-specific expression
• BulkTissue: analyses of public bulk brain RNA-seq from Parikshak et al., the GTEx consortium, and an autism spectrum disorder dataset found in Parikshak et al. Broadly corresponds to Figure 6.
• Other: miscellaneous analysis scripts. CrossTissue: deconvolution of heart and pancreas data.
• Fun: contains general custom functions for calling in other scripts

Datasets

The sequencing data generated in this study have been deposited in the GEO database under accession code GSE175772.

For generating signatures

Processed signature data can be accessed as Supplementary Data 5 in our study. Links to access the raw data underlying this can be found in the table below.

Name	Species	Description	Reference and Link	Used In
CA	Human	Post-mortem brain tissue, smartSeq2 snRNA-seq	Hodge, R. D. et al. Conserved cell types with divergent features in human versus mouse cortex. Nature 573, 61–68 (2019). Human Cell Atlas website, specifically this	-------
DM	Human	Surgically-resected brain tissue, smartSeq scRNA-seq	Darmanis, S. et al. A survey of human brain transcriptome diversity at the single cell level. Proc. Natl Acad. Sci. USA 112, 7285–7290 (2015). GEO, or processed version on GitHub	-------
F5	Human	Cultured brain cells, bulk CAGE-seq	Forrest, A. R. R. et al. A promoter-level mammalian expression atlas. Nature 507, 462–470 (2014). FANTOM5 Website, see ./Files for easier processed versions in which TSS peaks are pooled to genes	----
NG	Human	Post-mortem brain tissue, 10X snRNA-seq	Nagy, C. et al. Single-nucleus transcriptomics of the prefrontal cortex in major depressive disorder implicates oligodendrocyte precursor cells and excitatory neurons. Nat. Neurosci. 23, 771–781 (2020). GEO	-------
VL	Human	Post-mortem brain tissue, 10X snRNA-seq	Velmeshev, D. et al. Single-cell genomics identifies cell type–specific molecular changes in autism. Science (80-.) 364, 685–689 (2019). Data browser, use to access files 1 and 2	-------
LK	Human	Post-mortem brain tissue, 10X snRNA-seq	Lake, B. B. et al. Integrative single-cell analysis of transcriptional and epigenetic states in the human adult brain. Nat. Biotechnol. 36, 70–80 (2018). GEO, specifically this please	----
IP	Human	Surgically-resected brain tissue, bulk RNA-seq	Zhang, Y. et al. Purification and characterization of progenitor and mature human astrocytes reveals transcriptional and functional differences with mouse. Neuron 89, 37–53 (2016). Available on GEO, but use Table S4	-------
MM	Mouse	Mouse brain, bulk RNA-seq	Zhang, Y. et al. An RNA-sequencing transcriptome and splicing database of glia, neurons, and vascular cells of the cerebral cortex. J. Neurosci. 34, 11929–11947 (2014). Paper website (We note that this website is now down, and provide the original downloaded file in ./Files/Zhang2014_mouseRNASeq_originalDownload.xlsx)	----
TS	Mouse	Mouse brain, SmartSeq2 scRNA-seq	Tasic, B. et al. Shared and distinct transcriptomic cell types across neocortical areas. Nature 563, 72–78 (2018). GEO, with metadata in Supplementary Table 10	-------

Bulk RNA-seq datasets deconvolved as exemplars

Name	Samples	Details	Reference and Link
Parikshak	Human	Ribo-depleted RNA-seq from 251 post-mortem samples including frontal cortex, temporal cortex, and cerebellar vermis samples from 48 ASD and 49 control individuals, aged 2–67	Parikshak, N. N. et al. Genome-wide changes in lncRNA, splicing, and regional gene expression patterns in autism. Nature 540, 423–427 (2016). GitHub, specifically this RDA, including metadata
GTEx V7 Brain	Human	Poly-A+ RNA-seq of 1671 post-mortem samples, including 13 subregions which we classified as cortical, sub-cortical, cerebellar, or spinal	Consortium, Gte. Genetic effects on gene expression across human tissues. Nature 550, 204–213 (2017). GTEx website, specifically V7 counts, and two metadata files A, and B

Non-brain data

We also generated cell-type-specific signatures for two non-brain tissues, to test the tissue-specificity of our results.

Name	Tissue	Description	Reference and Link	Used In
EN	Pancreas	FACS-isolated tissue, single-cell RNA-seq	Enge, M. et al. Single-cell analysis of human pancreas reveals transcriptional signatures of aging and somatic mutation patterns. Cell 171, 321–330e14 (2017). GEO	----
BL	Pancreas	FACS-isolated tissue, bulk RNA-seq	Blodgett, D. M. et al. Novel observations from next-generation RNA sequencing of highly purified human adult and fetal islet cell subsets. Diabetes 64, 3172–3181 (2015). GEO	-------
FS	Pancreas	FACS-isolated tissue, bulk RNA-seq	Furuyama, K. et al. Diabetes relief in mice by glucose-sensing insulin-secreting human α-cells. Nature 567, 43–48 (2019). GEO	-------
FG	Pancreas	Cultured FACS-isolated tissue, bulk RNA-seq	Furuyama, K. et al. Diabetes relief in mice by glucose-sensing insulin-secreting human α-cells. Nature 567, 43–48 (2019). GEO	-------
F5	Heart	Cultured cells, bulk CAGE-seq	Forrest, A. R. R. et al. A promoter-level mammalian expression atlas. Nature 507, 462–470 (2014). FANTOM5 Website, see ./Files for easier processed versions in which TSS peaks are pooled to genes	-------
EN	Heart	Cultured cells, RNA-seq	Djebali, S. et al. Landscape of transcription in human cells. Nature 489, 101–108 (2012). ENCODE	-------
SC	Heart	Fresh tissue, scRNA-seq using iCell8	Wang, L. et al. Single-cell reconstruction of the adult human heart during heart failure and recovery reveals the cellular landscape underlying cardiac function. Nat. Cell Biol. 22, 108–119 (2020). GEO	-------

Deconvolution Algorithms

Method	Classification	Reference and installation
DeconRNASeq	Partial deconvolution	Gong, T. & Szustakowski, J. D. DeconRNASeq: a statistical framework for deconvolution of heterogeneous tissue samples based on mRNA-Seq data. Bioinformatics 29, 1083–1085 (2013). Bioconductor
CIBERSORT	Partial deconvolution	Newman, A. M. et al. Robust enumeration of cell subsets from tissue expression profiles. Nat. Methods 12, 453–457 (2015). Note: the R source for this package is available by request at the CIBERSORT website
dtangle	Partial deconvolution	Hunt, G. J., Freytag, S., Bahlo, M. & Gagnon-Bartsch, J. A. dtangle: accurate and robust cell type deconvolution. Bioinformatics 290262 (2018). CRAN
MuSiC	Partial deconvolution	Wang, X., Park, J., Susztak, K., Zhang, N. R. & Li, M. Bulk tissue cell type deconvolution with multi-subject single-cell expression reference. Nat. Commun. 10, 380 (2019). GitHub
Linseed	Complete deconvolution	Zaitsev, K., Bambouskova, M., Swain, A. & Artyomov, M. N. Complete deconvolution of cellular mixtures based on linearity of transcriptional signatures. Nat. Commun. 10, 2209 (2019). GitHub
Coex	Complete enrichment	Kelley, K. W., Nakao-Inoue, H., Molofsky, A. V. & Oldham, M. C. Variation among intact tissue samples reveals the core transcriptional features of human CNS cell classes. Nat. Neurosci. 21, 265397 (2018). Our own implementation is provided in the scripts.
BrainInABlender	Enrichment	Hagenauer, M. H. et al. Inference of cell type content from human brain transcriptomic datasets illuminates the effects of age, manner of death, dissection, and psychiatric diagnosis. PLoS ONE 13, 89391 (2018). GitHub
xCell	Enrichment	Aran, D., Hu, Z. & Butte, A. J. xCell: digitally portraying the tissue cellular heterogeneity landscape. Genome Biol. 18, 1–14 (2017). GitHub

Other packages

For core analyses

In progress

For auxiliary work, e.g. data wrangling and visualisation

In progress

Directory structure

These scripts assume the below directory structure. Feel free to rename the root as best suits you, however the scripts use ./BrainCellularComposition_GitHub. Please ave all downloaded files into ./Data/Raw

./BrainCellularComposition_GitHub
├── Data
│   ├── Preprocessed
│   │   └── CIB
│   └── Raw
├── Results
│   ├── ConfoundingComposition
│   │   ├── CompositionOnly
│   │   ├── CompositionAndExpression
│   │   └── CIBERSORTx
│   ├── BulkTissue
│   │   ├── Autism
│   │   ├── GTEx
│   │   └── Parikshak
│   ├── Other
│   │   ├── RNAvsCellContent
│   │   └── NonBrain
│   ├── BenchmarkingDatasets
│   │   ├── CA
│   │   ├── DM
│   │   ├── Supplementary
│   │   └── VL
└── Scripts

Other recommended readings

Deconvolution benchmarking in non-brain tissues, algorithm choice and data normalisation:

Avila Cobos, F., Alquicira-Hernandez, J., Powell, J.E. et al. Benchmarking of cell type deconvolution pipelines for transcriptomics data. Nat Commun 11, 5650 (2020). https://doi.org/10.1038/s41467-020-19015-1

Jin, H., Liu, Z. A benchmark for RNA-seq deconvolution analysis under dynamic testing environments. Genome Biol 22, 102 (2021). https://doi.org/10.1186/s13059-021-02290-6

Mohammadi, S., Zuckerman, N., Goldsmith, A., & Grama, A. A critical survey of deconvolution methods for separating cell types in complex tissues. Proceedings of the IEEE, 105(2), 340-366 (2017). https://doi.org/10.1109/JPROC.2016.2607121.