Transcriptome-wide prediction of prostate cancer gene expression from histopathology images using co-expression-based convolutional neural networks.

{"sci_version": "1.2", "schema": "https://corpus.bntrg.co.uk/sci/1.0/sci.schema.json", "id": "pmid:35595235", "kind": "science-record", "revision": 3, "content_hash": "sha256:ea77bc0edad48c45b266b63c2546ab99b7019eac73c1f785e03c732b01afe10c", "created_at": "2026-11-01T18:14:29Z", "updated_at": "2026-11-01T21:27:54Z", "generator": {"name": "corpus-assay-engine", "run_id": "reading-sci-2026-11-01"}, "paper": {"state": "complete", "title": "Transcriptome-wide prediction of prostate cancer gene expression from histopathology images using co-expression-based convolutional neural networks.", "doi": "10.1093/bioinformatics/btac343", "pmid": "35595235", "authors": ["Philippe Weitz", "Yinxi Wang", "Kimmo Kartasalo", "Lars Egevad", "Johan Lindberg", "Henrik Grönberg", "Martin Eklund", "Mattias Rantalainen"], "journal": "Bioinformatics (Oxford, England)", "year": 2022, "abstract": "MOTIVATION: Molecular phenotyping by gene expression profiling is central in contemporary cancer research and in molecular diagnostics but remains resource intense to implement. Changes in gene expression occurring in tumours cause morphological changes in tissue, which can be observed on the microscopic level. The relationship between morphological patterns and some of the molecular phenotypes can be exploited to predict molecular phenotypes from routine haematoxylin and eosin-stained whole slide images (WSIs) using convolutional neural networks (CNNs). In this study, we propose a new, computationally efficient approach to model relationships between morphology and gene expression.\nRESULTS: We conducted the first transcriptome-wide analysis in prostate cancer, using CNNs to predict bulk RNA-sequencing estimates from WSIs for 370 patients from the TCGA PRAD study. Out of 15 586 protein coding transcripts, 6618 had predicted expression significantly associated with RNA-seq estimates (FDR-adjusted P-value <1×10-4) in a cross-validation and 5419 (81.9%) of these associations were subsequently validated in a held-out test set. We furthermore predicted the prognostic cell-cycle progression score directly from WSIs. These findings suggest that contemporary computer vision models offer an inexpensive and scalable solution for prediction of gene expression phenotypes directly from WSIs, providing opportunity for cost-effective large-scale research studies and molecular diagnostics.\nAVAILABILITY AND IMPLEMENTATION: A self-contained example is available from http://github.com/phiwei/prostate_coexpression. Model predictions and metrics are available from doi.org/10.5281/zenodo.4739097.\nSUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.", "claims": [{"id": "claim-method", "statement": "Spearman rank correlations between the slide-level predictions and the RNA-seq expression level values were used as the primary performance metric.", "kind": "methodological", "target": null, "evidence": [{"asset": "source:pmid:35595235", "locator": "sections[6] \"2.4 Model optimization and performance evaluation\" paragraph 5", "quote": "Spearman rank correlations between the slide-level predictions and the RNA-seq expression level values were used as the primary performance metric."}], "state": "complete"}, {"id": "claim-data-access", "statement": "WSIs, RNA-seq and clinical data for the TCGA PRAD cohort are publicly available from the GDC data portal (https://portal.gdc.cancer.gov/). Information on biochemical recurrence was obtained from xenahubs\n(https://xenabrowser.net/datapages/?dataset=TCGA.PRAD.sampleMap/PRAD_clinicalMatrix&host=https://tcga.xenahubs.net). All patient-level expression predictions and computed correlations, as well as adjusted P-values, are available from https://zenodo.org/record/4739097#.YouJcu5Bwfs.", "kind": "methodological", "target": null, "evidence": [{"asset": "source:pmid:35595235", "locator": "sections[20] \"Data availability\" paragraphs 1-3", "quote": "WSIs, RNA-seq and clinical data for the TCGA PRAD cohort are publicly available from the GDC data portal (https://portal.gdc.cancer.gov/). Information on biochemical recurrence was obtained from xenahubs\n(https://xenabrowser.net/datapages/?dataset=TCGA.PRAD.sampleMap/PRAD_clinicalMatrix&host=https://tcga.xenahubs.net). All patient-level expression predictions and computed correlations, as well as adjusted P-values, are available from https://zenodo.org/record/4739097#.YouJcu5Bwfs."}], "state": "complete"}, {"id": "claim-target-result", "statement": "Out of 15 586 protein coding transcripts, 6618 had predicted expression significantly associated with RNA-seq estimates (FDR-adjusted P-value <1×10−4) in a cross-validation and 5419 (81.9%) of these associations were subsequently validated in a held-out test set.", "kind": "quantitative", "target": {"value": 81.9, "unit": "per cent", "tolerance": null, "source": "Out of 15 586 protein coding transcripts, 6618 had predicted expression significantly associated with RNA-seq estimates (FDR-adjusted P-value <1×10−4) in a cross-validation and 5419 (81.9%) of these associations were subsequently validated in a held-out test set."}, "evidence": [{"asset": "source:pmid:35595235", "locator": "sections[0] \"Abstract\" paragraph 2", "quote": "Out of 15 586 protein coding transcripts, 6618 had predicted expression significantly associated with RNA-seq estimates (FDR-adjusted P-value <1×10−4) in a cross-validation and 5419 (81.9%) of these associations were subsequently validated in a held-out test set."}], "state": "complete"}], "gaps": []}, "experiment": {"state": "partial", "objective": "Out of 15 586 protein coding transcripts, 6618 had predicted expression significantly associated with RNA-seq estimates (FDR-adjusted P-value <1×10−4) in a cross-validation and 5419 (81.9%) of these associations were subsequently validated in a held-out test set.", "steps": [{"id": "read_published_estimate", "order": 1, "name": "Read the published estimate from the source", "kind": "other", "state": "complete", "blocked_on": null, "description": "The claim, method and evidence span published with the source article.", "duration_hours": null, "parameters": {}, "inputs": ["source:pmid:35595235"], "outputs": ["the published estimate and its evidence"], "instruments": [], "reagents": []}, {"id": "obtain_source_data", "order": 2, "name": "Obtain the source data from the named repository", "kind": "data-preparation", "state": "blocked", "blocked_on": "data:accession_absent", "description": "The GEO series accession resolves and its sample accessions are listed. This lane did not download the dataset or read its licence.", "duration_hours": null, "parameters": {}, "inputs": [], "outputs": ["the source data with its repository provenance"], "instruments": [], "reagents": []}, {"id": "run_analysis", "order": 3, "name": "Run the published analysis", "kind": "analysis", "state": "complete", "blocked_on": null, "description": "Apply the paper's statistical method to the resolved series and compare the result with the published target.", "duration_hours": null, "parameters": {}, "inputs": [], "outputs": ["the replicated result and its comparison"], "instruments": [], "reagents": []}], "materials": [], "instruments": [], "reagents": [], "conditions": {}, "gaps": ["step:obtain_source_data blocked on data:accession_absent"]}, "data": {"state": "partial", "assets": [{"id": "source:pmid:35595235", "role": "manuscript", "format": "xml", "media_type": "application/xml", "uri": "https://doi.org/10.1093/bioinformatics/btac343", "sha256": "sha256:c809aa404b22de71566f06c2c3f5c5ccffe8b59c2a1c2d7d9c2e74f1e11ed766", "bytes": 50704, "access": "open", "state": "resolved", "reader": null, "reader_status": "reader_required", "identifier": "10.1093/bioinformatics/btac343", "version": null, "note": "Full article fetched from PMC; the checksum covers the rendered section body, not the original XML bytes.", "structure": {"pmcid": "PMC9237721", "source": "europe_pmc_fulltext", "coverage": "whole_body", "text_hash_scope": "rendered_section_body"}}], "gaps": []}, "analysis": {"state": "complete", "methods": [{"id": "method-1", "family": "correlation_analysis", "specific": "spearman", "label": "Correlation analysis", "dsl_operations": ["spearman_rho"], "missing_operations": [], "expressible": true, "parameters": {}, "data_shapes": ["paired_numeric_columns"], "evidence": {"score": 1.0, "band": "verbatim", "quote": "Spearman rank correlations between the slide-level predictions and the RNA-seq expression level values were used as the primary performance metric."}}], "targets": [{"id": "target-published-estimate", "claim_id": "claim-target-result", "operation": null, "expected_value": 81.9, "unit": "per cent", "tolerance": null, "state": "complete", "note": "The published estimate as the reader quoted it from the paper. No tolerance is published because the paper states none."}, {"id": "target-published-estimate-p-value", "claim_id": "claim-target-result", "operation": null, "expected_value": 1.0, "unit": "p-value", "tolerance": null, "state": "complete", "note": "The threshold the paper applied to its own result, as the paper states it. A run that reproduces the effect is judged against this threshold."}], "gaps": []}, "capability": {"state": "complete", "requirements": [{"capability_id": "capability:computational_analysis", "label": "Computational analysis platform", "assay_families": ["meta_analysis", "machine_learning", "survival_analysis", "statistical_regression", "pca", "molecular_docking", "network_analysis", "differential_expression", "gwas"], "availability_state": "owned", "tier_unlocked": "R1", "candidate": "Automated Assays Laravel application and simulated adapters"}], "gaps": []}, "readiness": {"state": "blocked", "tier": "R4", "blocking_axis": "data", "blocked_on": "data:accession_absent", "axes": {"method": {"tier": "R1", "expressible": true, "required_operations": ["spearman_rho"], "missing_operations": [], "basis": "The paper names the method it used and the corpus's method extractor maps it onto the DSL catalogue: spearman."}, "capability": {"tier": "R1", "capability_id": "capability:computational_analysis", "availability_state": "owned", "basis": "The computational analysis is owned. The physical bench the paper used is not, so a physical validation cannot be run here."}, "data": {"tier": "R4", "reachable": false, "accessions": [], "resolved_accessions": [], "statements": ["WSIs, RNA-seq and clinical data for the TCGA PRAD cohort are publicly available from the GDC data portal (https://portal.gdc.cancer.gov/). Information on biochemical recurrence was obtained from xenahubs\n(https://xenabrowser.net/datapages/?dataset=TCGA.PRAD.sampleMap/PRAD_clinicalMatrix&host=https://tcga.xenahubs.net). All patient-level expression predictions and computed correlations, as well as adjusted P-values, are available from https://zenodo.org/record/4739097#.YouJcu5Bwfs."], "basis": "The paper names no accession to its source data, so the data has to come from the authors."}}, "can_run": {"dsl_operations": ["spearman_rho"], "assets": ["source:pmid:35595235"], "steps": ["run_analysis"]}, "must_build": {"dsl_operations": [], "converters": [], "readers": [], "capabilities": [], "data_gaps": ["data:geo:None"]}, "gaps": ["data:accession_absent"]}, "review": {"state": "partial", "machine_critique": {"kind": "extraction_review", "reader": {"model": "MiniMax-M3.1-Flash", "provider": "minimax", "provider_family": "minimax", "read_at": "2026-11-01T17:46:50Z", "independence": "single_reader"}, "evidence_check": {"method": "verbatim", "availability": "verbatim"}, "calls": 5, "usage": {"prompt_tokens": 13284, "completion_tokens": 2014, "cached_tokens": 264}}, "human_critique": null, "stages": [{"id": "extraction", "state": "complete", "reader": "MiniMax-M3.1-Flash", "provider": "minimax"}], "gaps": ["review.independent_reader", "review.machine_critique", "review.human_critique"]}, "execution": {"state": "absent", "status": "not_run", "mode": null, "runs": [], "gaps": ["execution.not_run"]}, "provenance": {"entities": [{"id": "pmid:35595235", "type": "paper", "label": "pmid:35595235"}, {"id": "source:pmid:35595235", "type": "manuscript", "label": "PMID 35595235 full text"}, {"id": "data:geo:None", "type": "dataset", "label": "GEO None"}], "activities": [{"id": "activity-reading-sci-2026-11-01", "type": "classification", "used": [{"id": "pmid:35595235", "type": "paper", "label": "pmid:35595235"}, {"id": "source:pmid:35595235", "type": "manuscript", "label": "PMID 35595235 full text"}, {"id": "data:geo:None", "type": "dataset", "label": "GEO None"}], "agent": "agent:corpus-assay-engine"}, {"id": "activity-read-35595235", "type": "extraction", "used": ["source:pmid:35595235"], "agent": "agent:minimax-reader"}, {"id": "activity-resolve-None", "type": "accession_resolution", "used": ["data:geo:None"], "agent": "agent:corpus-assay-engine"}], "agents": [{"id": "agent:corpus-assay-engine", "type": "software", "label": "Corpus assay engine"}, {"id": "agent:minimax-reader", "type": "software", "label": "MiniMax literature reader"}], "source_run_id": "reading-sci-2026-11-01", "dsl_catalogue_fingerprint": "343fb30756ba6ec3", "capability_table_fingerprint": "2cb3e42f047d2ecf", "snapshot_id": "ba34b49490d9bfaf"}, "lineage": {"parent_id": "pmid:35595235", "parent_hash": "sha256:dee842da1f01c43c07dc3560f36aea2c51cadb3c43f20a95d5acf42b795f3baf", "fork_of": null, "generation": 2}, "gaps": ["execution.not_run", "review.human_critique", "review.independent_reader", "review.machine_critique", "step:obtain_source_data blocked on data:accession_absent"], "assays": [{"role": "data_generation", "definition_id": "rna_seq", "definition_version": "99e1195c7d6f", "definition_digest": null, "required_operations": [], "support_state": "partial", "source_hashes": {"assay_lexicon": "99e1195c7d6f"}, "confidence": 0.99, "evidence": ["Transcriptome-wide prediction of prostate cancer gene expr", "prostate cancer, using CNNs to predict bulk RNA-sequencing estimates from WSIs for 370 patients from th", "ted expression significantly associated with RNA-seq estimates (FDR-adjusted P-value <1×10-4) in", "expression. RESULTS: We conducted the first transcriptome-wide analysis in prostate cancer, using CNNs"], "order_basis": "paper_text_order:0", "id": "assay:1"}, {"role": "analysis", "definition_id": "correlation_analysis", "definition_version": "f2132e96ae6c7a4e", "definition_digest": null, "required_operations": ["spearman_rho"], "support_state": "supported", "source_hashes": {"dsl_catalogue": "f2132e96ae6c7a4e"}, "confidence": 1.0, "evidence": ["Spearman rank correlations between the slide-level predictions and the RNA-seq expression level values were used as the primary performance metric."], "order_basis": "consumes_earlier_output", "id": "assay:2"}]}