{"artifact_id":"L1-520","layer":"L1","title":"RNA-seq Cell-Type Classification (PWDR)","domain":"Computational Biology","sub_domain":"Single-cell / bulk RNA-seq transcript quantification with marker-gene-panel cell-type categorical readout","physics_fingerprint":{"L_DAG":8.3,"title":"RNA-seq Cell-Type Classification (PWDR)","domain":"Computational Biology","carrier":"rna_molecules_with_sequencing","primitives":["L.poly_a_capture","L.reverse_transcription","L.pcr_amplification","L.sequencing","L.transcriptome_alignment","L.transcript_quantification","L.normalization","L.marker_panel_classifier","int.cell"],"sub_domain":"Single-cell / bulk RNA-seq transcript quantification with marker-gene-panel cell-type categorical readout","noise_model":"negative_binomial","problem_class":"linear_inverse_with_categorical_readout","solution_space":"1D_celltype_label","difficulty_delta":5,"integration_axis":"molecular_cellular","sensing_mechanism":"rnaseq_with_marker_panel_classifier"},"observable_profile":{"metric":"categorical_accuracy","regime":"Existence guaranteed within Omega bounds. Uniqueness conditional on adequate sequencing depth (typically >30k reads per cell for 3' chemistry) and adequate marker-panel coverage. Stability conditional with dropout_rate dominant for low-expressing markers; batch_effect dominant cross-sample; doublet_contamination dominant for high cell densities. Joint Hadamard well-posedness for the coupled RNA-seq + marker-panel-classifier forward established by Trapnell 2014 (foundational scRNA-seq), Macosko 2015 (Drop-seq), Stuart-Butler 2019 (Seurat v3 integration), Tabula Sapiens Consortium 2022, Zhang 2019 (CellMarker), Franzen 2019 (PanglaoDB).","secondary":"macro_F1_per_celltype"},"size_tiers":{"center_spec":{"omega":{"N_cells":5000,"N_genes":20000,"batch_effect":0.0,"dropout_rate":0.0,"reads_per_cell":50000,"sequencing_depth":50000,"library_complexity":0.85,"doublet_contamination":0.0,"taxonomy_disagreement":0.0,"ambient_rna_contamination":0.0,"marker_panel_coverage_uncertainty":0.0},"input_format":"scrnaseq_count_matrix_with_celltype_label","problem_class":"scrnaseq_celltype_pwdr","forward_operator":"scrnaseq_celltype_pwdr_forward","epsilon_fn_center":"0.85_accuracy"},"omega_bounds":{"N_cells":[100,1000000],"N_genes":[1000,60000],"dropout_rate":[0.0,0.5],"reads_per_cell":[1000,1000000]},"epsilon_bounds":{"categorical_accuracy":[0.4,0.99]},"allowed_problem_classes":["scrnaseq_celltype_pwdr","scrnaseq_immune_celltype_pwdr","scrnaseq_epithelial_pwdr","scrnaseq_neural_pwdr","bulk_rnaseq_deconvolution_pwdr"],"allowed_omega_dimensions":["N_cells","N_genes","reads_per_cell","sequencing_depth","library_complexity","dropout_rate","batch_effect","doublet_contamination","ambient_rna_contamination","marker_panel_coverage_uncertainty","taxonomy_disagreement"],"allowed_forward_operators":["scrnaseq_celltype_pwdr_forward","bulk_rnaseq_deconvolution_pwdr_forward","scrnaseq_immune_pwdr_forward","scrnaseq_atlas_pwdr_forward"]},"hardness_fn":{"type":"epsilon_fn","delta":5,"kappa":200,"metric":"categorical_accuracy"},"initiator_dataset":[{"name":"primary","weight":1.0,"ipfs_cid":null,"license_hash":null}],"status":"testnet","staked_pwm":0.0,"chain_hash":"0x673b55bb788633e2b747d4088754fce2dd908b8e80fb87a580eae551bba1a5d1","chain_tx_hash":"0xc076736d4f2f15c82c65bb5a3b13c16628e630d165c72e056fb413fb169ed1eb","chain_block":41555318}