From 3d4e6833c1d4ac118dbf1d8a12503db224b62eff Mon Sep 17 00:00:00 2001 From: Robrecht Cannoodt Date: Tue, 28 Jul 2026 14:03:16 +0200 Subject: [PATCH 1/3] correct the logcounts type file_dataset_sp.yaml declared `logcounts` as integer. Log-transformed counts are doubles, and the shipped datasets store them as such. --- src/api/file_dataset_sp.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/api/file_dataset_sp.yaml b/src/api/file_dataset_sp.yaml index ffb6a077..c97c9811 100644 --- a/src/api/file_dataset_sp.yaml +++ b/src/api/file_dataset_sp.yaml @@ -13,7 +13,7 @@ info: name: counts description: Raw counts required: true - - type: integer + - type: double name: logcounts description: Log-transformed counts required: true From d4ba07b2afc8f711567b498fdde3e003fb0e3ce8 Mon Sep 17 00:00:00 2001 From: Robrecht Cannoodt Date: Tue, 28 Jul 2026 14:16:43 +0200 Subject: [PATCH 2/3] give the downstream metrics real bounds and labels * clustering_ari was declared -Inf..+Inf. The adjusted Rand index is at most 1 and does not go below -0.5, so the report had nothing to rescale against. * ctdeconvolute_rmse was 0..+Inf, but it is the RMSE between two vectors of proportions, so it cannot exceed 1. * The eight labels were the metric ids repeated back; the ks metrics already had readable ones. --- src/metrics/downstream/config.vsh.yaml | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-) diff --git a/src/metrics/downstream/config.vsh.yaml b/src/metrics/downstream/config.vsh.yaml index c9668a2e..ce76e725 100644 --- a/src/metrics/downstream/config.vsh.yaml +++ b/src/metrics/downstream/config.vsh.yaml @@ -5,7 +5,7 @@ name: downstream info: metrics: - name: clustering_ari - label: clustering_ari + label: Adjusted Rand index summary: Adjusted rand index (ARI) measures the similarity between two clusters in real and simulated datasets. description: | Adjusted Rand Index used in spatial clustering to measure the similarity between two data clusterings, adjusted for chance. @@ -14,11 +14,11 @@ info: links: documentation: https://cran.r-project.org/web/packages/aricode/index.html repository: https://github.com/jchiquet/aricode - min: -Inf - max: +Inf + min: -1 + max: 1 maximize: true - name: clustering_nmi - label: clustering_nmi + label: Normalised mutual information summary: Normalized mutual information (NMI) measures of the mutual dependence between the real and simulated spatial clusters. description: | Normalized Mutual Information used in spatial clustering to measure the agreement between two different clusterings, scaled to [0, 1]. @@ -31,7 +31,7 @@ info: max: 1 maximize: true - name: svg_recall - label: svg_recall + label: SVG recall summary: Recall measures the proportion of real SVG correctly identified in the simulated dataset. description: | Recall used in identifying spatial variable genes, measuring the true positive rate. @@ -41,7 +41,7 @@ info: max: 1 maximize: true - name: svg_precision - label: svg_precision + label: SVG precision summary: Precision measures the proportion of correctly identified items in simulated datasets. description: | Precision used in identifying spatial variable genes, measuring the accuracy of positive predictions. @@ -51,7 +51,7 @@ info: max: 1 maximize: true - name: ctdeconvolute_rmse - label: ctdeconvolute_rmse + label: Cell type deconvolution RMSE summary: Root Mean Square deviation is calculated between the true and predicted proportion of per cell type. description: | Root Mean Squared Error used in cell type deconvolution to measure the difference between observed and predicted values. @@ -61,10 +61,10 @@ info: documentation: https://cran.r-project.org/web/packages/Metrics/Metrics.pdf repository: https://github.com/mfrasco/Metrics min: 0 - max: +Inf + max: 1 maximize: false - name: ctdeconvolute_jsd - label: ctdeconvolute_jsd + label: Cell type deconvolution JSD summary: Jensen-Shannon divergence (JSD) is calculated between the true and predicted proportion per cell type in all spots. description: | Jensen-Shannon Divergence used in cell type deconvolution to measure the similarity between two probability distributions. @@ -77,7 +77,7 @@ info: max: 1 maximize: false - name: crosscor_mantel - label: crosscor_mantel + label: Mantel statistic summary: Mantel statistic is the test statistic for the Mantel test, which is a correlation coefficient calculated between bivariate Moran’s I of real dataset and that of in simulation dataset. description: | Mantel statistic used in spatial cross-correlation to test the correlation between two distance matrices. @@ -90,7 +90,7 @@ info: max: 1 maximize: true - name: crosscor_cosine - label: crosscor_cosine + label: Cosine similarity summary: Cosine similarity measures similarity between bivariate Moran’s I of real dataset and that of in simulation dataset. description: | Cosine similarity used in spatial cross-correlation to measure the cosine of the angle between two non-zero vectors. From c0265f89d933c93e871df178d87b7f1959826b84 Mon Sep 17 00:00:00 2001 From: Robrecht Cannoodt Date: Tue, 28 Jul 2026 14:19:45 +0200 Subject: [PATCH 3/3] update changelog --- CHANGELOG.md | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/CHANGELOG.md b/CHANGELOG.md index e3468069..d4b2ca1c 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -1,6 +1,10 @@ # task_spatial_simulators dev Bug fixes: + - `file_dataset_sp.yaml`: `logcounts` is a double, not an integer. + - `downstream`: `clustering_ari` was declared -Inf..+Inf and + `ctdeconvolute_rmse` 0..+Inf, though both are bounded. Metric labels were + the ids repeated back. - `run_benchmark`: raise `uns_length_cutoff` from 15 to 50, so that `extract_uns_metadata` no longer drops the `metric_ids` of components that emit more than 15 metrics. All 28 `ks_statistic_gene_cell` metrics were