Preparing Counts and Metadata for VISTA
VISTA Development Team
Source:vignettes/guides/VISTA-input-preparation.Rmd
VISTA-input-preparation.RmdOverview
VISTA expects two main inputs for create_vista():
- A count table with one gene identifier column plus one column per sample.
- A sample metadata table (
sample_info) containingsample_namesand the grouping variables used in the analysis.
In practice, users often start from count tables whose column names are file paths, alignment outputs, or sequencing-derived sample identifiers. Metadata can also be incomplete or missing entirely. VISTA’s input preparation helpers are designed to standardize these objects before differential analysis starts:
Load Example Data
library(VISTA)
library(dplyr)
library(tibble)
data("count_data", package = "VISTA")
data("sample_metadata", package = "VISTA")
dim(count_data)
#> [1] 63677 9
head(sample_metadata[, c("sample_names", "cond_long")])
#> # A tibble: 6 × 2
#> sample_names cond_long
#> <chr> <chr>
#> 1 SRR1039508 control
#> 2 SRR1039509 treatment1
#> 3 SRR1039512 control
#> 4 SRR1039513 treatment1
#> 5 SRR1039516 control
#> 6 SRR1039517 treatment11. Standardize Counts
For an ordinary in-memory count table,
read_vista_counts() keeps the existing API contract and
returns a create_vista()-ready structure.
prepared_counts <- read_vista_counts(
count_data[seq_len(50), ],
format = "matrix",
gene_id_column = "gene_id"
)
names(prepared_counts)
#> [1] "counts" "row_data" "column_geneid" "sample_names"
#> [5] "sample_name_map" "input_format" "report"
head(prepared_counts$counts[, seq_len(4)])
#> gene_id SRR1039508 SRR1039509 SRR1039512
#> 1 ENSG00000000003 679 448 873
#> 2 ENSG00000000005 0 0 0
#> 3 ENSG00000000419 467 515 621
#> 4 ENSG00000000457 260 211 263
#> 5 ENSG00000000460 60 55 40
#> 6 ENSG00000000938 0 0 2
head(prepared_counts$row_data)
#> gene_id
#> ENSG00000000003 ENSG00000000003
#> ENSG00000000005 ENSG00000000005
#> ENSG00000000419 ENSG00000000419
#> ENSG00000000457 ENSG00000000457
#> ENSG00000000460 ENSG00000000460
#> ENSG00000000938 ENSG00000000938The returned object includes:
-
counts: standardized count table withgene_id -
row_data: feature metadata aligned to the counts -
sample_names: standardized sample columns -
sample_name_map: mapping from original to repaired sample names
2. Repair File-Derived Sample Names
RNA-seq count tables often inherit sample columns from alignment or quantification files. VISTA can repair these names conservatively when they are clearly file-derived and the repaired names remain unique.
counts_paths <- count_data[seq_len(20), c("gene_id", sample_metadata$sample_names[seq_len(4)]), drop = FALSE]
colnames(counts_paths)[2:5] <- c(
"/proj/run/03_alignment/HET_1_U/HET_1_U_star_alignAligned.sortedByCoord.out.bam",
"/proj/run/03_alignment/HET_1_ovary/HET_1_ovary_star_alignAligned.sortedByCoord.out.bam",
"/proj/run/03_alignment/WT_1_U/WT_1_U_star_alignAligned.sortedByCoord.out.bam",
"/proj/run/03_alignment/WT_1_ovary/WT_1_ovary_star_alignAligned.sortedByCoord.out.bam"
)
prepared_paths <- read_vista_counts(
counts_paths,
format = "matrix",
gene_id_column = "gene_id",
repair_sample_names = "auto"
)
prepared_paths$sample_name_map
#> original
#> 1 /proj/run/03_alignment/HET_1_U/HET_1_U_star_alignAligned.sortedByCoord.out.bam
#> 2 /proj/run/03_alignment/HET_1_ovary/HET_1_ovary_star_alignAligned.sortedByCoord.out.bam
#> 3 /proj/run/03_alignment/WT_1_U/WT_1_U_star_alignAligned.sortedByCoord.out.bam
#> 4 /proj/run/03_alignment/WT_1_ovary/WT_1_ovary_star_alignAligned.sortedByCoord.out.bam
#> repaired
#> 1 HET_1_U
#> 2 HET_1_ovary
#> 3 WT_1_U
#> 4 WT_1_ovaryAutomatic repair currently handles patterns such as:
- full file paths
- STAR outputs (
Aligned.sortedByCoord.out.bam,ReadsPerGene.out.tab) - RSEM outputs (
.genes.results,.isoforms.results) - generic quantification files such as
quant.sf - lane/read suffixes like
_S1_L001_R1_001
3. Derive Starter Metadata from Sample Names
If you do not yet have a metadata sheet,
derive_vista_metadata() can create a starter
sample_info table from the count sample names.
Split-based parsing
derived_split <- derive_vista_metadata(
prepared_paths,
parser = "split",
split = "_",
fields = c("genotype", "replicate", "tissue")
)
derived_split
#> sample_names genotype replicate tissue
#> 1 HET_1_U HET 1 U
#> 2 HET_1_ovary HET 1 ovary
#> 3 WT_1_U WT 1 U
#> 4 WT_1_ovary WT 1 ovaryRegex-based parsing
For sequencing accessions or other structured labels, regex parsing gives more control:
derived_regex <- derive_vista_metadata(
counts = NULL,
sample_names = c("SRR1039508", "SRR1039509", "SRR1039512"),
parser = "regex",
pattern = "SRR(\\d+)",
fields = "run_id"
)
derived_regex
#> sample_names run_id
#> 1 SRR1039508 1039508
#> 2 SRR1039509 1039509
#> 3 SRR1039512 1039512Template mode
Template mode adds empty placeholders for columns users commonly fill
in next, such as group and batch.
derive_vista_metadata(
counts = NULL,
sample_names = c("sampleA", "sampleB"),
parser = "none",
return_type = "template"
)
#> sample_names group batch
#> 1 sampleA <NA> <NA>
#> 2 sampleB <NA> <NA>4. Standardize Existing Metadata
If a metadata table already exists,
read_vista_metadata() standardizes it into the form
expected by create_vista().
prepared_samples <- read_vista_metadata(sample_metadata)
head(prepared_samples[, c("sample_names", "cond_long")])
#> sample_names cond_long
#> SRR1039508 SRR1039508 control
#> SRR1039509 SRR1039509 treatment1
#> SRR1039512 SRR1039512 control
#> SRR1039513 SRR1039513 treatment1
#> SRR1039516 SRR1039516 control
#> SRR1039517 SRR1039517 treatment1VISTA will infer sample_names from:
- an existing
sample_namescolumn - non-default rownames
- common aliases such as
sample,sample_id, orRun
5. Match Counts and Metadata
match_vista_inputs() aligns count columns and metadata
rows so they can be passed directly into
create_vista().
matched_inputs <- match_vista_inputs(prepared_counts, prepared_samples)
matched_inputs$column_geneid
#> [1] "gene_id"
identical(matched_inputs$sample_info$sample_names, colnames(matched_inputs$counts)[-1])
#> [1] TRUEWhen appropriate, you can drop unmatched samples instead of erroring:
sample_extra <- prepared_samples[seq_len(4), , drop = FALSE]
extra_row <- sample_extra[1, , drop = FALSE]
extra_row$sample_names <- "extra_sample"
if ("cond_long" %in% colnames(extra_row)) extra_row$cond_long <- "control"
if ("dex" %in% colnames(extra_row)) extra_row$dex <- "untrt"
if ("cell" %in% colnames(extra_row)) extra_row$cell <- "N999"
rownames(extra_row) <- "extra_sample"
sample_extra <- rbind(sample_extra, extra_row)
matched_drop <- match_vista_inputs(
prepared_counts,
sample_extra,
drop_unmatched = TRUE
)
matched_drop$report
#> $n_genes
#> [1] 50
#>
#> $n_samples
#> [1] 4
#>
#> $dropped_from_counts
#> [1] "SRR1039516" "SRR1039517" "SRR1039520" "SRR1039521"
#>
#> $dropped_from_sample_info
#> [1] "extra_sample"6. Create a VISTA Object
Once inputs are standardized, create_vista() is
unchanged.
vista <- create_vista(
counts = matched_inputs$counts,
sample_info = matched_inputs$sample_info,
column_geneid = matched_inputs$column_geneid,
group_column = "cond_long",
group_numerator = "treatment1",
group_denominator = "control",
method = "deseq2",
min_counts = 5,
min_replicates = 1
)
vista
#> class: VISTA
#> dim: 44 8
#> metadata(12): de_results de_summary ... design comparison
#> assays(1): norm_counts
#> rownames(44): ENSG00000000003 ENSG00000000419 ... ENSG00000004399
#> ENSG00000004455
#> rowData names(1): baseMean
#> colnames(8): SRR1039508 SRR1039509 ... SRR1039520 SRR1039521
#> colData names(14): SampleName cell ... sizeFactor sample_namesPractical Guidance
- Use
read_vista_counts()first whenever count columns come from external tools or file names. - Use
derive_vista_metadata()when you only have sample names and need a reviewable starter metadata sheet. - Use
read_vista_metadata()when you already have a metadata table and want VISTA to standardize it. - Use
match_vista_inputs()beforecreate_vista()to catch alignment issues early.
Session Info
sessionInfo()
#> R version 4.5.3 (2026-03-11)
#> Platform: x86_64-pc-linux-gnu
#> Running under: Ubuntu 24.04.4 LTS
#>
#> Matrix products: default
#> BLAS: /usr/lib/x86_64-linux-gnu/openblas-pthread/libblas.so.3
#> LAPACK: /usr/lib/x86_64-linux-gnu/openblas-pthread/libopenblasp-r0.3.26.so; LAPACK version 3.12.0
#>
#> locale:
#> [1] LC_CTYPE=C.UTF-8 LC_NUMERIC=C LC_TIME=C.UTF-8
#> [4] LC_COLLATE=C.UTF-8 LC_MONETARY=C.UTF-8 LC_MESSAGES=C.UTF-8
#> [7] LC_PAPER=C.UTF-8 LC_NAME=C LC_ADDRESS=C
#> [10] LC_TELEPHONE=C LC_MEASUREMENT=C.UTF-8 LC_IDENTIFICATION=C
#>
#> time zone: UTC
#> tzcode source: system (glibc)
#>
#> attached base packages:
#> [1] stats graphics grDevices utils datasets methods base
#>
#> other attached packages:
#> [1] tibble_3.3.1 dplyr_1.2.1 VISTA_0.99.8 BiocStyle_2.38.0
#>
#> loaded via a namespace (and not attached):
#> [1] RColorBrewer_1.1-3 jsonlite_2.0.0
#> [3] tidydr_0.0.6 magrittr_2.0.5
#> [5] ggtangle_0.1.1 farver_2.1.2
#> [7] rmarkdown_2.31 fs_2.0.1
#> [9] ragg_1.5.2 vctrs_0.7.3
#> [11] memoise_2.0.1 ggtree_4.0.5
#> [13] rstatix_0.7.3 htmltools_0.5.9
#> [15] S4Arrays_1.10.1 curl_7.0.0
#> [17] broom_1.0.12 Formula_1.2-5
#> [19] SparseArray_1.10.10 gridGraphics_0.5-1
#> [21] sass_0.4.10 bslib_0.10.0
#> [23] htmlwidgets_1.6.4 desc_1.4.3
#> [25] plyr_1.8.9 cachem_1.1.0
#> [27] igraph_2.2.3 lifecycle_1.0.5
#> [29] pkgconfig_2.0.3 gson_0.1.0
#> [31] Matrix_1.7-4 R6_2.6.1
#> [33] fastmap_1.2.0 MatrixGenerics_1.22.0
#> [35] digest_0.6.39 aplot_0.2.9
#> [37] enrichplot_1.30.5 colorspace_2.1-2
#> [39] ggnewscale_0.5.2 GGally_2.4.0
#> [41] patchwork_1.3.2 AnnotationDbi_1.72.0
#> [43] S4Vectors_0.49.1-1 DESeq2_1.50.2
#> [45] textshaping_1.0.5 GenomicRanges_1.62.1
#> [47] RSQLite_2.4.6 ggpubr_0.6.3
#> [49] polyclip_1.10-7 httr_1.4.8
#> [51] abind_1.4-8 compiler_4.5.3
#> [53] withr_3.0.2 bit64_4.6.0-1
#> [55] fontquiver_0.2.1 backports_1.5.1
#> [57] S7_0.2.1 BiocParallel_1.44.0
#> [59] carData_3.0-6 DBI_1.3.0
#> [61] ggstats_0.13.0 ggforce_0.5.0
#> [63] R.utils_2.13.0 ggsignif_0.6.4
#> [65] MASS_7.3-65 rappdirs_0.3.4
#> [67] DelayedArray_0.36.1 tools_4.5.3
#> [69] otel_0.2.0 scatterpie_0.2.6
#> [71] ape_5.8-1 msigdbr_26.1.0
#> [73] R.oo_1.27.1 glue_1.8.0
#> [75] nlme_3.1-168 GOSemSim_2.36.0
#> [77] grid_4.5.3 cluster_2.1.8.2
#> [79] reshape2_1.4.5 fgsea_1.36.2
#> [81] generics_0.1.4 gtable_0.3.6
#> [83] R.methodsS3_1.8.2 tidyr_1.3.2
#> [85] data.table_1.18.2.1 utf8_1.2.6
#> [87] car_3.1-5 XVector_0.50.0
#> [89] BiocGenerics_0.56.0 ggrepel_0.9.8
#> [91] pillar_1.11.1 stringr_1.6.0
#> [93] babelgene_22.9 limma_3.66.0
#> [95] yulab.utils_0.2.4 splines_4.5.3
#> [97] tweenr_2.0.3 treeio_1.34.0
#> [99] lattice_0.22-9 bit_4.6.0
#> [101] tidyselect_1.2.1 fontLiberation_0.1.0
#> [103] GO.db_3.22.0 locfit_1.5-9.12
#> [105] Biostrings_2.78.0 knitr_1.51
#> [107] fontBitstreamVera_0.1.1 bookdown_0.46
#> [109] IRanges_2.44.0 Seqinfo_1.0.0
#> [111] edgeR_4.8.2 SummarizedExperiment_1.40.0
#> [113] stats4_4.5.3 xfun_0.57
#> [115] Biobase_2.70.0 statmod_1.5.1
#> [117] matrixStats_1.5.0 stringi_1.8.7
#> [119] lazyeval_0.2.3 ggfun_0.2.0
#> [121] yaml_2.3.12 evaluate_1.0.5
#> [123] codetools_0.2-20 gdtools_0.5.0
#> [125] qvalue_2.42.0 BiocManager_1.30.27
#> [127] ggplotify_0.1.3 cli_3.6.6
#> [129] systemfonts_1.3.2 jquerylib_0.1.4
#> [131] Rcpp_1.1.1 png_0.1-9
#> [133] parallel_4.5.3 assertthat_0.2.1
#> [135] pkgdown_2.2.0 ggplot2_4.0.2
#> [137] blob_1.3.0 clusterProfiler_4.18.4
#> [139] DOSE_4.4.0 tidytree_0.4.7
#> [141] ggiraph_0.9.6 scales_1.4.0
#> [143] purrr_1.2.2 crayon_1.5.3
#> [145] rlang_1.2.0 cowplot_1.2.0
#> [147] fastmatch_1.1-8 KEGGREST_1.50.0