{"doi":"10.1093/bioinformatics/btae343","title":"HE2Gene: image-to-RNA translation via multi-task learning for spatial transcriptomics data","abstract":"<jats:title>Abstract</jats:title>\n               <jats:sec>\n                  <jats:title>Motivation</jats:title>\n                  <jats:p>Tissue context and molecular profiling are commonly used measures in understanding normal development and disease pathology. In recent years, the development of spatial molecular profiling technologies (e.g. spatial resolved transcriptomics) has enabled the exploration of quantitative links between tissue morphology and gene expression. However, these technologies remain expensive and time-consuming, with subsequent analyses necessitating high-throughput pathological annotations. On the other hand, existing computational tools are limited to predicting only a few dozen to several hundred genes, and the majority of the methods are designed for bulk RNA-seq.</jats:p>\n               </jats:sec>\n               <jats:sec>\n                  <jats:title>Results</jats:title>\n                  <jats:p>In this context, we propose HE2Gene, the first multi-task learning-based method capable of predicting tens of thousands of spot-level gene expressions along with pathological annotations from H&amp;E-stained images. Experimental results demonstrate that HE2Gene is comparable to state-of-the-art methods and generalizes well on an external dataset without the need for re-training. Moreover, HE2Gene preserves the annotated spatial domains and has the potential to identify biomarkers. This capability facilitates cancer diagnosis and broadens its applicability to investigate gene-disease associations.</jats:p>\n               </jats:sec>\n               <jats:sec>\n                  <jats:title>Availability and implementation</jats:title>\n                  <jats:p>The source code and data information has been deposited at https://github.com/Microbiods/HE2Gene.</jats:p>\n               </jats:sec>","journal":"Bioinformatics","year":2024,"id":630800,"datarank":0.32958368660043297,"base_score":2.1972245773362196,"endowment":2.1972245773362196,"self_citation_contribution":0.32958368660043297,"citation_network_contribution":0.0,"self_endowment_contribution":0.32958368660043297,"citer_contribution":0.0,"corpus_percentile":null,"corpus_rank":null,"citation_count":8,"citer_count":0,"citers_with_citation_signal":0,"citers_with_endowment":0,"datacite_reuse_total":0,"is_dataset":false,"is_dataset_confidence":null,"is_data_producer":false,"deposit_databanks":null,"is_oa":false,"file_count":0,"downloads":0,"has_version_chain":false,"published_date":null,"fair_score":null,"fair_percentile":null,"algorithm_id":"datarank_citation_only_1hop_v6","ranking_scope":"data_only","authors":[{"id":811240,"name":"Jiecong Lin","orcid":"0000-0003-3347-837X","position":1,"is_corresponding":false},{"id":1029249,"name":"Yuchen Wang","orcid":"0000-0002-5862-8322","position":2,"is_corresponding":false},{"id":552104,"name":"Weitong Zhang","orcid":"0000-0003-4731-9986","position":3,"is_corresponding":false},{"id":1634402,"name":"Weidun Xie","orcid":"0000-0001-9009-2647","position":4,"is_corresponding":false},{"id":580576,"name":"Zetian Zheng","orcid":"0000-0002-7199-0051","position":5,"is_corresponding":false},{"id":65476,"name":"Ka‐Chun Wong","orcid":"0000-0001-6062-733X","position":6,"is_corresponding":false},{"id":1160632,"name":"Xingjian Chen","orcid":"0000-0001-9639-1584","position":0,"is_corresponding":false}],"reference_count":0,"raw_metadata":{"has_enrichment":true,"resolved":true,"title":"HE2Gene: image-to-RNA translation via multi-task learning for spatial transcriptomics data","abstract":"<jats:title>Abstract</jats:title>\n               <jats:sec>\n                  <jats:title>Motivation</jats:title>\n                  <jats:p>Tissue context and molecular profiling are commonly used measures in understanding normal development and disease pathology. In recent years, the development of spatial molecular profiling technologies (e.g. spatial resolved transcriptomics) has enabled the exploration of quantitative links between tissue morphology and gene expression. However, these technologies remain expensive and time-consuming, with subsequent analyses necessitating high-throughput pathological annotations. On the other hand, existing computational tools are limited to predicting only a few dozen to several hundred genes, and the majority of the methods are designed for bulk RNA-seq.</jats:p>\n               </jats:sec>\n               <jats:sec>\n                  <jats:title>Results</jats:title>\n                  <jats:p>In this context, we propose HE2Gene, the first multi-task learning-based method capable of predicting tens of thousands of spot-level gene expressions along with pathological annotations from H&amp;E-stained images. Experimental results demonstrate that HE2Gene is comparable to state-of-the-art methods and generalizes well on an external dataset without the need for re-training. Moreover, HE2Gene preserves the annotated spatial domains and has the potential to identify biomarkers. This capability facilitates cancer diagnosis and broadens its applicability to investigate gene-disease associations.</jats:p>\n               </jats:sec>\n               <jats:sec>\n                  <jats:title>Availability and implementation</jats:title>\n                  <jats:p>The source code and data information has been deposited at https://github.com/Microbiods/HE2Gene.</jats:p>\n               </jats:sec>","is_dataset_classified":null,"base_score":2.1972245773362196,"endowment":2.1972245773362196,"datacite_reuse_total":0,"file_count":0,"downloads":0,"views":0,"has_version_chain":false,"is_dataset":false,"is_oa":false,"pmid":"38837395","pmcid":null,"openalex_id":"https://openalex.org/W4399369150","authors":[],"funders":[{"funder_name":"Ka-Chun Wong","grant_id":"","title":null},{"funder_name":"Ka-Chun Wong","grant_id":"","title":null}],"total_grants":2,"fwci":1.3171,"citation_percentile":0.79183005,"influential_citations":0,"citation_trend":[{"year":2024,"count":1},{"year":2025,"count":5},{"year":2026,"count":2}],"oa_status":"gold","license":"cc-by","oa_locations":[{"url":"https://academic.oup.com/bioinformatics/advance-article-pdf/doi/10.1093/bioinformatics/btae343/58108010/btae343.pdf","host_type":"journal"},{"url":"https://academic.oup.com/bioinformatics/advance-article-pdf/doi/10.1093/bioinformatics/btae343/58108010/btae343.pdf","host_type":"publisher"},{"url":"https://academic.oup.com/bioinformatics/advance-article-pdf/doi/10.1093/bioinformatics/btae343/58186422/btae343.pdf","host_type":"publisher"},{"url":"https://doi.org/10.1093/bioinformatics/btae343","host_type":"journal"},{"url":"https://pubmed.ncbi.nlm.nih.gov/38837395","host_type":"repository"},{"url":"https://www.ncbi.nlm.nih.gov/pmc/articles/11164830","host_type":"repository"},{"url":"https://hdl.handle.net/2031/1e88733e-bc5b-42a1-a29d-786b9617da66","host_type":"repository"},{"url":"https://pmc.ncbi.nlm.nih.gov/articles/PMC11164830/pdf/btae343.pdf","host_type":"repository"}],"fields_of_study":["Single-cell and spatial transcriptomics","Cell Image Analysis Techniques","Domain Adaptation and Few-Shot Learning","Humans","Transcriptome","Gene Expression Profiling","Computational Biology","Machine Learning","RNA"],"mesh_terms":["Machine Learning","Humans","RNA","Computational Biology","Gene Expression Profiling","Transcriptome"],"keywords":["Computer science","Profiling (computer programming)","Transcriptome","Context (archaeology)","Computational biology","Artificial intelligence","Spatial analysis","Spatial contextual awareness","Machine learning","Pattern recognition (psychology)","Gene","Gene expression","Biology","Genetics"],"sdg_mappings":[],"linked_datasets":[],"clinical_trials":[],"software_tools":[],"database_accessions":[],"source":"live","citation_network_status":"fetched"},"created_at":"2026-08-05T22:10:51.781457Z","pmid":null,"pmcid":null,"fwci":null,"citation_percentile":null,"influential_citations":0,"oa_status":null,"license":null,"views":0,"total_file_size_bytes":0,"version_count":0,"fair_f":null,"fair_a":null,"fair_i":null,"fair_r":null,"fair_zscore":null,"fair_rationale":null,"fair_model":null,"fair_agent_version":null,"fair_fulltext_source":null,"fair_has_llm":null,"fair_computed_at":null,"clinical_trials":[],"software_tools":[],"db_accessions":[],"linked_datasets":[],"topics":[]}