New embedding

test_submission/README.md

@@ -0,0 +1,18 @@
+# Embedding name: ESM-2
+
+- Modality: Amino acid sequence / protein sequence
+- Algorithm: Transformer protein language model
+- Source used here: Nobody knows
+- Embedding source URL: google.com
+- Original model/code URL: instagram.com
+
+## Public files
+
+- `embeddings.npz`: compressed NumPy archive containing the embedding matrix under key `embeddings`.
+- `genes.tsv`: row metadata table. Row order matches the embedding matrix.
+- `metadata.json`: machine-readable provenance and processing metadata.
+
+## Matrix
+
+- Shape: `18690 genes × 1280 dimensions`
+- Dtype: `float32`

test_submission/embeddings.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ce1d5a3ea8c8fe5ab5b7a7104cd4d3c31f615fff2d85f9eba54037261345a1be
+size 46833352

test_submission/metadata.json

@@ -0,0 +1,44 @@
+{
+  "embedding_name": "ESM2-v1.1",
+  "source_embedding": "Zhong_all_genes_ESM2",
+  "modality": "protein_sequence",
+  "algorithm": "ESM2 Transformer",
+  "source_identifier_type": "Entrez",
+  "mapping_strategy": "Exact Entrez ID mapping to master_gene_table_v1_1_enriched.csv. Only Entrez IDs mapping to exactly one master gene are accepted. Duplicate source rows mapping to the same final master gene are averaged.",
+  "counts": {
+    "source_rows": 18903,
+    "source_dimensions": 680,
+    "mapped_source_rows_before_duplicate_collapse": 18708,
+    "unmapped_source_rows": 35,
+    "ambiguous_source_rows": 160,
+    "final_mapped_genes": 18690,
+    "final_dimensions": 680,
+    "master_rows": 20097,
+    "duplicate_final_mappings_averaged": 16
+  },
+  "coverage": {
+    "source_row_mapping_fraction": 0.9896841771147437,
+    "master_gene_coverage_fraction": 0.9299895506792059
+  },
+  "embedding_id": "esm2",
+  "former_processing_embedding_id": "esm2_v1_1",
+  "public_file_structure": {
+    "embeddings": "embeddings.npz",
+    "genes": "genes.tsv",
+    "metadata": "metadata.json",
+    "readme": "README.md"
+  },
+  "num_genes": 18690,
+  "embedding_dimension": 680,
+  "matrix_shape": [
+    18690,
+    680
+  ],
+  "all_zero_rows": {
+    "count": 0,
+    "documented": true,
+    "expected": false,
+    "reason": "No exact all-zero rows were detected in embeddings.npz.",
+    "report_file": null
+  }
+}

test_submission/submission-template.yaml

@@ -0,0 +1,39 @@
+schema_version: 1
+embedding_id: Spongebob
+display_name: test_embedding
+  name: Michael
+  affiliation: ETH
+scientific:
+  # Broad catalog group. Recommended values: DNA, Expression,
+  # Functional Genomics, Knowledge, Literature, Multimodal,
+  # Network, Protein, RNA, Regulatory, or Single-cell.
+  # Other values are accepted but displayed under "Other".
+  category:  Protein
+  # Specific free-text description of the representation,
+  # e.g. protein sequence, single-cell transcriptomics,
+  # or regulatory activity.
+  modality: Replace with the modality
+  original_method: Replace with the model or generation method
+  model_checkpoint: Replace with checkpoint/version or not-applicable
+  source_publication: ''
+  description: Replace this text with a clear description of the embedding, how it
+    was generated, and its intended use.
+provenance:
+  embedding_generated_by: Replace with model, software, or authors
+  input_data_source: Replace with the original data source
+  input_data_release: Replace with version or release
+  species: Homo sapiens
+  pooling_strategy: Replace with pooling strategy or not-applicable
+  identifier_before_mapping: Replace with original identifier type
+  mapping_method: Replace with the identifier mapping procedure
+licensing:
+  source_license: MIT
+  redistribution_allowed: true
+  terms_url: google.com
+attestations:
+  information_is_accurate: true
+  rights_to_redistribute: true
+  no_sensitive_human_data: true
+  no_patient_level_data: true
+  source_terms_reviewed: true
+  acceptance_of_automatic_checks: true