Add Predict Modality task

CHANGELOG.md

@@ -30,6 +30,12 @@
 
 ## NEW CONTENT
 
+* Add Predict Modality benchmark page (PR #320).
+
+# openproblems.bio v2.3.6
+
+## NEW CONTENT
+
 * Add an event page for the Weekly wednesday work meeting (PR #299).
 
 * Add `Advanced_topics` pages to documentation (PR #300).

results/predict_modality/data/dataset_info.json

@@ -0,0 +1,68 @@
+[
+  {
+    "dataset_id": "openproblems_neurips2021/bmmc_cite/normal",
+    "dataset_name": "NeurIPS2021 CITE-Seq (GEX2ADT)",
+    "dataset_summary": "Single-cell CITE-Seq (GEX+ADT) data collected from bone marrow mononuclear cells of 12 healthy human donors.",
+    "dataset_description": "Single-cell CITE-Seq data collected from bone marrow mononuclear cells of 12 healthy human donors using the 10X 3 prime Single-Cell Gene Expression kit with Feature Barcoding in combination with the BioLegend TotalSeq B Universal Human Panel v1.0. The dataset was generated to support Multimodal Single-Cell Data Integration Challenge at NeurIPS 2021. Samples were prepared using a standard protocol at four sites. The resulting data was then annotated to identify cell types and remove doublets. The dataset was designed with a nested batch layout such that some donor samples were measured at multiple sites with some donors measured at a single site.",
+    "data_reference": "luecken2021neurips",
+    "data_url": "https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE194122",
+    "date_created": "25-11-2024",
+    "file_size": 704994,
+    "common_dataset_id": "openproblems_neurips2021/bmmc_cite"
+  },
+  {
+    "dataset_id": "openproblems_neurips2021/bmmc_multiome/normal",
+    "dataset_name": "NeurIPS2021 Multiome (GEX2ATAC)",
+    "dataset_summary": "Single-cell Multiome (GEX+ATAC) data collected from bone marrow mononuclear cells of 12 healthy human donors.",
+    "dataset_description": "Single-cell CITE-Seq data collected from bone marrow mononuclear cells of 12 healthy human donors using the 10X Multiome Gene Expression and Chromatin Accessibility kit. The dataset was generated to support Multimodal Single-Cell Data Integration Challenge at NeurIPS 2021. Samples were prepared using a standard protocol at four sites. The resulting data was then annotated to identify cell types and remove doublets. The dataset was designed with a nested batch layout such that some donor samples were measured at multiple sites with some donors measured at a single site.",
+    "data_reference": "luecken2021neurips",
+    "data_url": "https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE194122",
+    "date_created": "25-11-2024",
+    "file_size": 31080807,
+    "common_dataset_id": "openproblems_neurips2021/bmmc_multiome"
+  },
+  {
+    "dataset_id": "openproblems_neurips2021/bmmc_multiome/swap",
+    "dataset_name": "NeurIPS2021 Multiome (ATAC2GEX)",
+    "dataset_summary": "Single-cell Multiome (GEX+ATAC) data collected from bone marrow mononuclear cells of 12 healthy human donors.",
+    "dataset_description": "Single-cell CITE-Seq data collected from bone marrow mononuclear cells of 12 healthy human donors using the 10X Multiome Gene Expression and Chromatin Accessibility kit. The dataset was generated to support Multimodal Single-Cell Data Integration Challenge at NeurIPS 2021. Samples were prepared using a standard protocol at four sites. The resulting data was then annotated to identify cell types and remove doublets. The dataset was designed with a nested batch layout such that some donor samples were measured at multiple sites with some donors measured at a single site.",
+    "data_reference": "luecken2021neurips",
+    "data_url": "https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE194122",
+    "date_created": "25-11-2024",
+    "file_size": 7883109,
+    "common_dataset_id": "openproblems_neurips2021/bmmc_multiome"
+  },
+  {
+    "dataset_id": "openproblems_neurips2022/pbmc_cite/normal",
+    "dataset_name": "OpenProblems NeurIPS2022 CITE-Seq (GEX2ADT)",
+    "dataset_summary": "Single-cell CITE-Seq (GEX+ADT) data collected from bone marrow mononuclear cells of 12 healthy human donors.",
+    "dataset_description": "Single-cell CITE-Seq data collected from bone marrow mononuclear cells of 12 healthy human donors using the 10X 3 prime Single-Cell Gene Expression kit with Feature Barcoding in combination with the BioLegend TotalSeq B Universal Human Panel v1.0. The dataset was generated to support Multimodal Single-Cell Data Integration Challenge at NeurIPS 2022. Samples were prepared using a standard protocol at four sites. The resulting data was then annotated to identify cell types and remove doublets. The dataset was designed with a nested batch layout such that some donor samples were measured at multiple sites with some donors measured at a single site.",
+    "data_reference": "lance2024predicting",
+    "data_url": "https://www.kaggle.com/competitions/open-problems-multimodal/data",
+    "date_created": "25-11-2024",
+    "file_size": 591886,
+    "common_dataset_id": "openproblems_neurips2022/pbmc_cite"
+  },
+  {
+    "dataset_id": "openproblems_neurips2022/pbmc_cite/swap",
+    "dataset_name": "OpenProblems NeurIPS2022 CITE-Seq (ADT2GEX)",
+    "dataset_summary": "Single-cell CITE-Seq (GEX+ADT) data collected from bone marrow mononuclear cells of 12 healthy human donors.",
+    "dataset_description": "Single-cell CITE-Seq data collected from bone marrow mononuclear cells of 12 healthy human donors using the 10X 3 prime Single-Cell Gene Expression kit with Feature Barcoding in combination with the BioLegend TotalSeq B Universal Human Panel v1.0. The dataset was generated to support Multimodal Single-Cell Data Integration Challenge at NeurIPS 2022. Samples were prepared using a standard protocol at four sites. The resulting data was then annotated to identify cell types and remove doublets. The dataset was designed with a nested batch layout such that some donor samples were measured at multiple sites with some donors measured at a single site.",
+    "data_reference": "lance2024predicting",
+    "data_url": "https://www.kaggle.com/competitions/open-problems-multimodal/data",
+    "date_created": "25-11-2024",
+    "file_size": 32551804,
+    "common_dataset_id": "openproblems_neurips2022/pbmc_cite"
+  },
+  {
+    "dataset_id": "openproblems_neurips2021/bmmc_cite/swap",
+    "dataset_name": "NeurIPS2021 CITE-Seq (ADT2GEX)",
+    "dataset_summary": "Single-cell CITE-Seq (GEX+ADT) data collected from bone marrow mononuclear cells of 12 healthy human donors.",
+    "dataset_description": "Single-cell CITE-Seq data collected from bone marrow mononuclear cells of 12 healthy human donors using the 10X 3 prime Single-Cell Gene Expression kit with Feature Barcoding in combination with the BioLegend TotalSeq B Universal Human Panel v1.0. The dataset was generated to support Multimodal Single-Cell Data Integration Challenge at NeurIPS 2021. Samples were prepared using a standard protocol at four sites. The resulting data was then annotated to identify cell types and remove doublets. The dataset was designed with a nested batch layout such that some donor samples were measured at multiple sites with some donors measured at a single site.",
+    "data_reference": "luecken2021neurips",
+    "data_url": "https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE194122",
+    "date_created": "25-11-2024",
+    "file_size": 13467880,
+    "common_dataset_id": "openproblems_neurips2021/bmmc_cite"
+  }
+]

results/predict_modality/data/method_info.json

@@ -0,0 +1,130 @@
+[
+  {
+    "task_id": "control_methods",
+    "method_id": "mean_per_gene",
+    "method_name": "Mean per gene",
+    "method_summary": "Returns the mean expression value per gene.",
+    "method_description": "Returns the mean expression value per gene.",
+    "is_baseline": true,
+    "references_doi": null,
+    "references_bibtex": null,
+    "code_url": "https://github.com/openproblems-bio/task_predict_modality",
+    "documentation_url": null,
+    "image": "https://ghcr.io/openproblems-bio/task_predict_modality/control_methods/mean_per_gene:build_main",
+    "implementation_url": "https://github.com/openproblems-bio/task_predict_modality/blob/0bd597e201b39fbcbc1fcd7047f7654a9713a197/src/control_methods/mean_per_gene",
+    "code_version": "build_main",
+    "commit_sha": "0bd597e201b39fbcbc1fcd7047f7654a9713a197"
+  },
+  {
+    "task_id": "control_methods",
+    "method_id": "random_predict",
+    "method_name": "Random predictions",
+    "method_summary": "Returns random training profiles.",
+    "method_description": "Returns random training profiles.",
+    "is_baseline": true,
+    "references_doi": null,
+    "references_bibtex": null,
+    "code_url": "https://github.com/openproblems-bio/task_predict_modality",
+    "documentation_url": null,
+    "image": "https://ghcr.io/openproblems-bio/task_predict_modality/control_methods/random_predict:build_main",
+    "implementation_url": "https://github.com/openproblems-bio/task_predict_modality/blob/0bd597e201b39fbcbc1fcd7047f7654a9713a197/src/control_methods/random_predict",
+    "code_version": "build_main",
+    "commit_sha": "0bd597e201b39fbcbc1fcd7047f7654a9713a197"
+  },
+  {
+    "task_id": "control_methods",
+    "method_id": "zeros",
+    "method_name": "Zeros",
+    "method_summary": "Returns a prediction consisting of all zeros.",
+    "method_description": "Returns a prediction consisting of all zeros.",
+    "is_baseline": true,
+    "references_doi": null,
+    "references_bibtex": null,
+    "code_url": "https://github.com/openproblems-bio/task_predict_modality",
+    "documentation_url": null,
+    "image": "https://ghcr.io/openproblems-bio/task_predict_modality/control_methods/zeros:build_main",
+    "implementation_url": "https://github.com/openproblems-bio/task_predict_modality/blob/0bd597e201b39fbcbc1fcd7047f7654a9713a197/src/control_methods/zeros",
+    "code_version": "build_main",
+    "commit_sha": "0bd597e201b39fbcbc1fcd7047f7654a9713a197"
+  },
+  {
+    "task_id": "control_methods",
+    "method_id": "solution",
+    "method_name": "Solution",
+    "method_summary": "Returns the ground-truth solution.",
+    "method_description": "Returns the ground-truth solution.",
+    "is_baseline": true,
+    "references_doi": null,
+    "references_bibtex": null,
+    "code_url": "https://github.com/openproblems-bio/task_predict_modality",
+    "documentation_url": null,
+    "image": "https://ghcr.io/openproblems-bio/task_predict_modality/control_methods/solution:build_main",
+    "implementation_url": "https://github.com/openproblems-bio/task_predict_modality/blob/0bd597e201b39fbcbc1fcd7047f7654a9713a197/src/control_methods/solution",
+    "code_version": "build_main",
+    "commit_sha": "0bd597e201b39fbcbc1fcd7047f7654a9713a197"
+  },
+  {
+    "task_id": "methods",
+    "method_id": "knnr_py",
+    "method_name": "KNNR (Py)",
+    "method_summary": "K-nearest neighbor regression in Python.",
+    "method_description": "K-nearest neighbor regression in Python.",
+    "is_baseline": false,
+    "references_doi": "10.2307/1403797",
+    "references_bibtex": null,
+    "code_url": "https://github.com/openproblems-bio/task_predict_modality",
+    "documentation_url": null,
+    "image": "https://ghcr.io/openproblems-bio/task_predict_modality/methods/knnr_py:build_main",
+    "implementation_url": "https://github.com/openproblems-bio/task_predict_modality/blob/0bd597e201b39fbcbc1fcd7047f7654a9713a197/src/methods/knnr_py",
+    "code_version": "build_main",
+    "commit_sha": "0bd597e201b39fbcbc1fcd7047f7654a9713a197"
+  },
+  {
+    "task_id": "methods",
+    "method_id": "knnr_r",
+    "method_name": "KNNR (R)",
+    "method_summary": "K-nearest neighbor regression in R.",
+    "method_description": "K-nearest neighbor regression in R.",
+    "is_baseline": false,
+    "references_doi": "10.2307/1403797",
+    "references_bibtex": null,
+    "code_url": "https://github.com/openproblems-bio/task_predict_modality",
+    "documentation_url": null,
+    "image": "https://ghcr.io/openproblems-bio/task_predict_modality/methods/knnr_r:build_main",
+    "implementation_url": "https://github.com/openproblems-bio/task_predict_modality/blob/0bd597e201b39fbcbc1fcd7047f7654a9713a197/src/methods/knnr_r",
+    "code_version": "build_main",
+    "commit_sha": "0bd597e201b39fbcbc1fcd7047f7654a9713a197"
+  },
+  {
+    "task_id": "methods",
+    "method_id": "lm",
+    "method_name": "Linear Model",
+    "method_summary": "Linear model regression.",
+    "method_description": "A linear model regression method.",
+    "is_baseline": false,
+    "references_doi": "10.2307/2346786",
+    "references_bibtex": null,
+    "code_url": "https://github.com/openproblems-bio/task_predict_modality",
+    "documentation_url": null,
+    "image": "https://ghcr.io/openproblems-bio/task_predict_modality/methods/lm:build_main",
+    "implementation_url": "https://github.com/openproblems-bio/task_predict_modality/blob/0bd597e201b39fbcbc1fcd7047f7654a9713a197/src/methods/lm",
+    "code_version": "build_main",
+    "commit_sha": "0bd597e201b39fbcbc1fcd7047f7654a9713a197"
+  },
+  {
+    "task_id": "methods",
+    "method_id": "guanlab_dengkw_pm",
+    "method_name": "Guanlab-dengkw",
+    "method_summary": "A kernel ridge regression method with RBF kernel.",
+    "method_description": "This is a solution developed by Team Guanlab - dengkw in the Neurips 2021 competition to predict one modality\nfrom another using kernel ridge regression (KRR) with RBF kernel. Truncated SVD is applied on the combined\ntraining and test data from modality 1 followed by row-wise z-score normalization on the reduced matrix. The\ntruncated SVD of modality 2 is predicted by training a KRR model on the normalized training matrix of modality 1.\nPredictions on the normalized test matrix are then re-mapped to the modality 2 feature space via the right\nsingular vectors.\n",
+    "is_baseline": false,
+    "references_doi": "10.1101/2022.04.11.487796",
+    "references_bibtex": null,
+    "code_url": "https://github.com/openproblems-bio/task_predict_modality",
+    "documentation_url": null,
+    "image": "https://ghcr.io/openproblems-bio/task_predict_modality/methods/guanlab_dengkw_pm:build_main",
+    "implementation_url": "https://github.com/openproblems-bio/task_predict_modality/blob/0bd597e201b39fbcbc1fcd7047f7654a9713a197/src/methods/guanlab_dengkw_pm",
+    "code_version": "build_main",
+    "commit_sha": "0bd597e201b39fbcbc1fcd7047f7654a9713a197"
+  }
+]