Fix NULL row names in NMF matrix for sig_tally() Wang method

[Copilot]

When using sig_tally() with the Wang method, the NMF matrix row names are NULL, particularly for single-sample cases. The Steele method correctly preserves sample IDs as row names.

Root Cause

The matrix generation transposes a components-by-samples matrix to samples-by-components. R's transpose doesn't reliably preserve dimension names in edge cases (e.g., single-column matrices).

Changes

• R/sig_tally.R: Explicitly capture sample names before transpose and assign as row names after
• tests/testthat/test-sig_tally_rownames.R: Add tests for single-sample and multi-sample cases
• NEWS.md: Document bug fix

Example

# Single sample case (previously returned NULL row names)
cn <- read_copynumber(singleSampleSegTabs, ...)
tally_wang <- sig_tally(cn, method = "Wang")
rownames(tally_wang$nmf_matrix)  # Now returns: "TCGA-A8-A07S-01A-11D-A036-01"

The fix ensures consistent behavior across all sample counts and aligns with the Steele method's row name handling.

Original prompt

---

This section details on the original issue you should resolve

<issue_title>NULL Row Names in NMF Matrix When Using sig_tally() with Wang Method</issue_title>
<issue_description>Hi @ShixiangWang,

Hope you're well.

Just noticed that when using the sig_tally() function with the Wang method, the row names in the resulting NMF matrix (tally_wang$nmf_matrix) are unexpectedly NULL. This behavior is inconsistent with the expected outcome, where the row names would ideally be the sample IDs, as observed when using the S (Steele) method. Reprex below:

# Load the sigminer library
library(sigminer)
#> Registered S3 method overwritten by 'sigminer':
#>   method      from
#>   print.bytes Rcpp
#> sigminer version 2.3.2
#> - Star me at https://github.com/ShixiangWang/sigminer
#> - Run hello() to see usage and citation.

# Load a toy segmentation table included with the sigminer package
load(system.file("extdata", "toy_segTab.RData",
                 package = "sigminer", mustWork = TRUE
))

# Set a seed for reproducibility
set.seed(1234)

# Add a new column 'minor_cn' with random values of either 0 or 1
segTabs$minor_cn <- sample(c(0, 1), size = nrow(segTabs), replace = TRUE)

# Subset the segmentation table for a single sample
singleSampleSegTabs <- subset(segTabs, sample == "TCGA-A8-A07S-01A-11D-A036-01")

# Read the copy number data for the single sample
cn <- read_copynumber(singleSampleSegTabs,
                      seg_cols = c("chromosome", "start", "end", "segVal"),
                      genome_measure = "wg", complement = TRUE, add_loh = TRUE
)
#> ℹ [2024-09-05 14:58:13.990813]: Started.
#> ℹ [2024-09-05 14:58:13.995483]: Genome build  : hg19.
#> ℹ [2024-09-05 14:58:13.996023]: Genome measure: wg.
#> ℹ [2024-09-05 14:58:13.996513]: When add_loh is TRUE, use_all is forced to TRUE.
#> Please drop columns you don't want to keep before reading.
#> ✔ [2024-09-05 14:58:14.002538]: Chromosome size database for build obtained.
#> ℹ [2024-09-05 14:58:14.003157]: Reading input.
#> ✔ [2024-09-05 14:58:14.003672]: A data frame as input detected.
#> ✔ [2024-09-05 14:58:14.004327]: Column names checked.
#> ✔ [2024-09-05 14:58:14.005043]: Column order set.
#> ✔ [2024-09-05 14:58:14.006148]: Chromosomes unified.
#> ✔ [2024-09-05 14:58:14.008141]: Value 2 (normal copy) filled to uncalled chromosomes.
#> ✔ [2024-09-05 14:58:14.009723]: Data imported.
#> ℹ [2024-09-05 14:58:14.010245]: Segments info:
#> ℹ [2024-09-05 14:58:14.010774]:     Keep - 45
#> ℹ [2024-09-05 14:58:14.011272]:     Drop - 0
#> ✔ [2024-09-05 14:58:14.011938]: Segments sorted.
#> ℹ [2024-09-05 14:58:14.012445]: Adding LOH labels...
#> ℹ [2024-09-05 14:58:14.013251]: Joining adjacent segments with same copy number value. Be patient...
#> ✔ [2024-09-05 14:58:14.018461]: 39 segments left after joining.
#> ✔ [2024-09-05 14:58:14.019082]: Segmental table cleaned.
#> ℹ [2024-09-05 14:58:14.019575]: Annotating.
#> ✔ [2024-09-05 14:58:14.024584]: Annotation done.
#> ℹ [2024-09-05 14:58:14.02513]: Summarizing per sample.
#> ✔ [2024-09-05 14:58:14.033565]: Summarized.
#> ℹ [2024-09-05 14:58:14.034175]: Generating CopyNumber object.
#> ✔ [2024-09-05 14:58:14.034964]: Generated.
#> ℹ [2024-09-05 14:58:14.03548]: Validating object.
#> ✔ [2024-09-05 14:58:14.036014]: Done.
#> ℹ [2024-09-05 14:58:14.036588]: 0.046 secs elapsed.

# Tally signatures using the Wang method
tally_wang <- sigminer::sig_tally(cn, method = "Wang", keep_only_matrix = FALSE)
#> ℹ [2024-09-05 14:58:14.03747]: Started.
#> ℹ [2024-09-05 14:58:14.038619]: Step: getting copy number features.
#> ℹ [2024-09-05 14:58:14.076979]: Getting breakpoint count per 10 Mb...
#> ℹ [2024-09-05 14:58:14.104202]: Getting breakpoint count per chromosome arm...
#> ℹ [2024-09-05 14:58:14.111208]: Getting copy number...
#> ℹ [2024-09-05 14:58:14.112797]: Getting change-point copy number change...
#> ℹ [2024-09-05 14:58:14.118523]: Getting length of chains of oscillating copy number...
#> ℹ [2024-09-05 14:58:14.124469]: Getting (log10 based) segment size...
#> ℹ [2024-09-05 14:58:14.125993]: Getting the minimal number of chromosome with 50% CNV...
#> ℹ [2024-09-05 14:58:14.133015]: Getting burden of chromosome...
#> ✔ [2024-09-05 14:58:14.140732]: Gotten.
#> ℹ [2024-09-05 14:58:14.14139]: Step: generating copy number components.
#> ✔ [2024-09-05 14:58:14.141898]: `feature_setting` checked.
#> ℹ [2024-09-05 14:58:14.143294]: Step: counting components.
#> ✔ [2024-09-05 14:58:14.250663]: Counted.
#> ℹ [2024-09-05 14:58:14.251405]: Step: generating components by sample matrix.
#> ✔ [2024-09-05 14:58:14.252393]: Matrix generated.
#> ℹ [2024-09-05 14:58:14.252925]: 0.215 secs elapsed.

# Tally signatures using the Steele method
tally_steele <- sigminer::sig_tally(cn, method = "S", keep_only_matrix = FALSE)
#> ℹ [2024-09-05 14:58:14.253776]: Started.
#> ℹ [2024-09-05 14:58:14.25461]...

</details>



<!-- START COPILOT CODING AGENT SUFFIX -->

- Fixes ShixiangWang/sigminer#465

<!-- START COPILOT CODING AGENT TIPS -->
---

✨ Let Copilot coding agent [set things up for you](https://github.com/ShixiangWang/sigminer/issues/new?title=✨+Set+up+Copilot+instructions&body=Configure%20instructions%20for%20this%20repository%20as%20documented%20in%20%5BBest%20practices%20for%20Copilot%20coding%20agent%20in%20your%20repository%5D%28https://gh.io/copilot-coding-agent-tips%29%2E%0A%0A%3COnboard%20this%20repo%3E&assignees=copilot) — coding agent works faster and does higher quality work when set up for your repo.