Wei-Sheng Wu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Wei-Sheng Wu has a total of 185 co-authors. Here are the top co-authors by number of shared publications:
- Tzu-Hsien Yang
Department of Biomedical Engineering, National Cheng Kung University, University Road, 701 Tainan, Taiwan; Medical Device Innovation Center, National Cheng Kung University, Tainan City 701, Taiwan
14 shared publications Yan-Yuan Tseng
Center for Molecular Medicine and Genetics, Wayne State University, School of Medicine, Detroit, Michigan 48201, United States
7 shared publications- Wen-Hsiung Li
Biodiversity Research Center, Academia Sinica
6 shared publications - Yan-Yuan Tseng
Center for Molecular Medicine and Genetics, Wayne State University, School of Medicine, Detroit, MI 48201, USA
6 shared publications Fu-Jou Lai
Department of Electrical Engineering, National Cheng Kung University, Tainan, Taiwan
5 shared publicationsChia-Chun Chiu
Department of Electrical Engineering, National Cheng Kung University, No.1 University Road, 70101, Tainan, Taiwan
4 shared publications
Latest Publications
- 22 October 2026
- Published by Elsevier in Journal of Molecular Biology
- Vol. 439 (1) , 170074
piRTarBase2: An Updated C. elegans piRNA Binding Site Database With Extensive Evidence for piRNA Targeting Events
- 22 October 2026
- Published by Elsevier in Journal of Molecular Biology
- Vol. 439 (1) , 170074
piRTarBase2: An Updated C. elegans piRNA Binding Site Database With Extensive Evidence for piRNA Targeting Events
- Open Access
- 10 June 2026
- Published by American Chemical Society (ACS) in Environmental Science & Technology
- Vol. 60 (24) , 17341-17354
Enhanced Environmental PFAS Characterization Using a Virtual High-Resolution Mass Spectral Library Generated by Transfer Learning-Based Neural Network
Per- and polyfluoroalkyl substances (PFAS) represent a critical class of persistent environmental contaminants with significant ecological and human health implications. However, the rapid emergence of novel PFAS has far outpaced the development of reference mass spectral databases. Here, Neural Per- and Polyfluoroalkyl Substances Mass Spectrometry (NPFAS-MS), a transfer learning-based neural network model, was developed to predict PFAS-specific high-resolution mass spectra. NPFAS-MS was fine-tuned from a pretrained model using PFAS tandem mass (MS/MS) spectra. NPFAS-MS outperformed other in silico spectral prediction models for PFAS spectra prediction across multiple spectral similarity metrics. In library searching tasks, libraries generated by other spectral prediction models showed top-1 recall between 42.1% and 55.4%, while NPFAS-MS demonstrated 71.1%. Applying the virtual PFAS mass spectral library generated with NPFAS-MS using 10,553 PFAS structures from the U.S. EPA and NORMAN databases to groundwater and aqueous film-forming foam (AFFF) samples revealed more potential PFAS than other mass spectral databases. Specifically, 38 potential PFAS were annotated in AFFF products and 40 in groundwater samples. NPFAS-MS enabled characterization of emerging PFAS, including ultrashort-chain, unsaturated, and substituted derivatives in environmental matrices. This advancement enables comprehensive environmental monitoring of rapidly evolving PFAS contamination. NPFAS-MS and associated resources were deployed as a web-based tool at https://cosbi10.ee.ncku.edu.tw/NPFAS_MS/, enabling both structure-to-spectrum prediction and library searching against 31,659 predicted PFAS spectra.
- 10 June 2026
- Published by American Chemical Society (ACS) in Environmental Science & Technology
- Vol. 60 (24) , 17341-17354
Open AccessEnhanced Environmental PFAS Characterization Using a Virtual High-Resolution Mass Spectral Library Generated by Transfer Learning-Based Neural Network
Per- and polyfluoroalkyl substances (PFAS) represent a critical class of persistent environmental contaminants with significant ecological and human health implications. However, the rapid emergence of novel PFAS has far outpaced the development of reference mass spectral databases. Here, Neural Per- and Polyfluoroalkyl Substances Mass Spectrometry (NPFAS-MS), a transfer learning-based neural network model, was developed to predict PFAS-specific high-resolution mass spectra. NPFAS-MS was fine-tuned from a pretrained model using PFAS tandem mass (MS/MS) spectra. NPFAS-MS outperformed other in silico spectral prediction models for PFAS spectra prediction across multiple spectral similarity metrics. In library searching tasks, libraries generated by other spectral prediction models showed top-1 recall between 42.1% and 55.4%, while NPFAS-MS demonstrated 71.1%. Applying the virtual PFAS mass spectral library generated with NPFAS-MS using 10,553 PFAS structures from the U.S. EPA and NORMAN databases to groundwater and aqueous film-forming foam (AFFF) samples revealed more potential PFAS than other mass spectral databases. Specifically, 38 potential PFAS were annotated in AFFF products and 40 in groundwater samples. NPFAS-MS enabled characterization of emerging PFAS, including ultrashort-chain, unsaturated, and substituted derivatives in environmental matrices. This advancement enables comprehensive environmental monitoring of rapidly evolving PFAS contamination. NPFAS-MS and associated resources were deployed as a web-based tool at https://cosbi10.ee.ncku.edu.tw/NPFAS_MS/, enabling both structure-to-spectrum prediction and library searching against 31,659 predicted PFAS spectra.
- 15 October 2025
- Published by Elsevier in Journal of Molecular Biology
- Vol. 437 (24) , 169488
NoAC: an automatic builder for knowledge bases and query interfaces on genomes of non-model organisms
- 15 October 2025
- Published by Elsevier in Journal of Molecular Biology
- Vol. 437 (24) , 169488
NoAC: an automatic builder for knowledge bases and query interfaces on genomes of non-model organisms
- Open Access
- 3 April 2025
- Published by Oxford University Press (OUP) in Database: The Journal of Biological Databases and Curation
- Vol. 2025
mirTarCLASH: a comprehensive miRNA target database based on chimeric read-based experiments
MicroRNAs (miRNAs) can target messenger RNAs to control their degradation or translation repression effects. Therefore, identifying the target and binding sites of different miRNAs is essential for understanding miRNA functions. To investigate these interactions, researchers have employed the cross-linking, ligation, and sequencing of hybrids (CLASH-seq) and similar CLASH-like approaches to generate chimeric reads formed by miRNAs and their targeting segments. These chimeric reads allow for the direct extraction of both the miRNA–target gene pairs and their corresponding binding sites. Nevertheless, these studies lack user-friendly platforms for researchers to investigate these interactions efficiently, thus hindering scientists’ ability to explore miRNA functions. To address this gap, we developed mirTarCLASH, a comprehensive database that deposits 502 061/322 707/224 452 unique hybrid reads from human/mouse/worm miRNA chimeric read-based experiments. In mirTarCLASH, the chimera analysis algorithm ChiRA and two distinct binding site inference tools, RNAup and miRanda, were adopted to facilitate the exploration of miRNA–target pairs derived from CLASH-like experiments. Compared with existing similar repositories, mirTarCLASH further enables several confidence evaluation filters with visualization functions for the extracted results. The results can be further refined based on the key properties of the miRNA targeting sites, including read depths, numbers of supporting algorithms, and cross-linking-induced mutations, to enhance confidence levels. In addition, these miRNA-binding sites are visually represented through an integrated transcript atlas. Finally, we demonstrated the biological applicability of mirTarCLASH via the well-characterized example interaction between cel-let-7-5p and lin-41 in Caenorhabditis elegans, showcasing the potential of mirTarCLASH to provide novel insights for subsequent experimental research designs. The constructed mirTarCLASH database is freely available at https://cosbi.ee.ncku.edu.tw/MirTarClash. Database URL: https://cosbi.ee.ncku.edu.tw/MirTarClash
- 3 April 2025
- Published by Oxford University Press (OUP) in Database: The Journal of Biological Databases and Curation
- Vol. 2025
Open AccessmirTarCLASH: a comprehensive miRNA target database based on chimeric read-based experiments
MicroRNAs (miRNAs) can target messenger RNAs to control their degradation or translation repression effects. Therefore, identifying the target and binding sites of different miRNAs is essential for understanding miRNA functions. To investigate these interactions, researchers have employed the cross-linking, ligation, and sequencing of hybrids (CLASH-seq) and similar CLASH-like approaches to generate chimeric reads formed by miRNAs and their targeting segments. These chimeric reads allow for the direct extraction of both the miRNA–target gene pairs and their corresponding binding sites. Nevertheless, these studies lack user-friendly platforms for researchers to investigate these interactions efficiently, thus hindering scientists’ ability to explore miRNA functions. To address this gap, we developed mirTarCLASH, a comprehensive database that deposits 502 061/322 707/224 452 unique hybrid reads from human/mouse/worm miRNA chimeric read-based experiments. In mirTarCLASH, the chimera analysis algorithm ChiRA and two distinct binding site inference tools, RNAup and miRanda, were adopted to facilitate the exploration of miRNA–target pairs derived from CLASH-like experiments. Compared with existing similar repositories, mirTarCLASH further enables several confidence evaluation filters with visualization functions for the extracted results. The results can be further refined based on the key properties of the miRNA targeting sites, including read depths, numbers of supporting algorithms, and cross-linking-induced mutations, to enhance confidence levels. In addition, these miRNA-binding sites are visually represented through an integrated transcript atlas. Finally, we demonstrated the biological applicability of mirTarCLASH via the well-characterized example interaction between cel-let-7-5p and lin-41 in Caenorhabditis elegans, showcasing the potential of mirTarCLASH to provide novel insights for subsequent experimental research designs. The constructed mirTarCLASH database is freely available at https://cosbi.ee.ncku.edu.tw/MirTarClash. Database URL: https://cosbi.ee.ncku.edu.tw/MirTarClash
- Open Access
- 2 January 2025
- Published by Springer Nature in BMC Plant Biology
- Vol. 25 (1) , 1-13
OrchidBase 6.0: increasing the number of Cymbidium (Orchidaceae) genomes and new bioinformatic tools for orchid genome analysis
Background: Orchids are well-known for their rich diversity of species as well as wide range habitats. Their floral structures are so unique in angiosperms that many of orchids are economically and culturally important in human society. Orchids pollination strategy and evolutionary trajectory are also fantastic human for centuries. Previously, OrchidBase was created not only for storage and management of orchid genomic and transcriptomic information including Apostasia shenzhenica, Dendrobium catenatum, Phalaenopsis equestris, and two species of Platanthera that belong to three different subfamilies of Orchidaceae, but explored orchid genetic sequences for their function. The OrchidBase offers an opportunity for the plant science community to compare orchid genomes and transcriptomes, and retrieve orchid sequences for further study. Description: Recently, three whole-genome sequences of the Epidendroideae species, Cymbidium sinense, C. ensifolium and C. goeringii, were sequenced de novo, assembled, and analyzed. In addition, the systemic transcriptomes of these three species have been established. We included these datasets to develop a new version of OrchidBase 6.0. Furthermore, four new analytical methods, namely regulation, updated transcriptome, advanced BLAST, and domain search, were developed for orchid genome analyses. Conclusion: OrchidBase 6.0 extended genetic information to that of eight orchid species and created new tools for an expanded community curation in response to the ever-increasing volume and complexity of data.
- 2 January 2025
- Published by Springer Nature in BMC Plant Biology
- Vol. 25 (1) , 1-13
Open AccessOrchidBase 6.0: increasing the number of Cymbidium (Orchidaceae) genomes and new bioinformatic tools for orchid genome analysis
Background: Orchids are well-known for their rich diversity of species as well as wide range habitats. Their floral structures are so unique in angiosperms that many of orchids are economically and culturally important in human society. Orchids pollination strategy and evolutionary trajectory are also fantastic human for centuries. Previously, OrchidBase was created not only for storage and management of orchid genomic and transcriptomic information including Apostasia shenzhenica, Dendrobium catenatum, Phalaenopsis equestris, and two species of Platanthera that belong to three different subfamilies of Orchidaceae, but explored orchid genetic sequences for their function. The OrchidBase offers an opportunity for the plant science community to compare orchid genomes and transcriptomes, and retrieve orchid sequences for further study. Description: Recently, three whole-genome sequences of the Epidendroideae species, Cymbidium sinense, C. ensifolium and C. goeringii, were sequenced de novo, assembled, and analyzed. In addition, the systemic transcriptomes of these three species have been established. We included these datasets to develop a new version of OrchidBase 6.0. Furthermore, four new analytical methods, namely regulation, updated transcriptome, advanced BLAST, and domain search, were developed for orchid genome analyses. Conclusion: OrchidBase 6.0 extended genetic information to that of eight orchid species and created new tools for an expanded community curation in response to the ever-increasing volume and complexity of data.
- Open Access
- 1 January 2025
- Published by American Association for the Advancement of Science (AAAS) in Computational and Structural Biotechnology Journal
- Vol. 27, 2832-2842
Identifying the risk of Kawasaki disease based solely on routine blood test features through novel construction of machine learning models
Kawasaki disease (KD) is a leading cause of acquired coronary vasculitis in children and remains a critical diagnostic challenge among febrile pediatric patients. To support frontline pediatricians with a more objective diagnostic tool, we developed and implemented KDpredictor, a machine learning-based model for KD risk identification. KDpredictor leverages only the routine blood test features, including complete blood count with differential count, C-reactive protein, and alanine aminotransferase. It also takes the lead in using age-calibrated eosinophil, platelet, and hemoglobin results. Trained using the light gradient boosting machine algorithm on clinical data from 1,927 KD cases and 45,274 febrile controls, KDpredictor achieved strong performance metrics (auROC: 95.7%, auPRC: 72.4%, recall: 0.89) on a reserved test set, outperforming previous models by at least 3% in auROC and 39.3% in auPRC. Additional explainable AI analyses revealed that several top predictive features in KDpredictor are consistent with prior clinical findings. We also evaluated KDpredictor on three independent cohorts collected in East Asia (Taiwan and China) during the COVID-19 period. KDpredictor achieves recall values of 90.9%, 83.7%, and 91.7% on KD samples identified in three independent medical centers, respectively, indicating its applicability across independent clinical settings. In summary, KDpredictor demonstrates robust generalizability in KD risk identification across populations by using only standard blood samples independent of clinical symptoms. KDpredictor is freely available at https://cosbi.ee.ncku.edu.tw/KD_under7/.
- 1 January 2025
- Published by American Association for the Advancement of Science (AAAS) in Computational and Structural Biotechnology Journal
- Vol. 27, 2832-2842
Open AccessIdentifying the risk of Kawasaki disease based solely on routine blood test features through novel construction of machine learning models
Kawasaki disease (KD) is a leading cause of acquired coronary vasculitis in children and remains a critical diagnostic challenge among febrile pediatric patients. To support frontline pediatricians with a more objective diagnostic tool, we developed and implemented KDpredictor, a machine learning-based model for KD risk identification. KDpredictor leverages only the routine blood test features, including complete blood count with differential count, C-reactive protein, and alanine aminotransferase. It also takes the lead in using age-calibrated eosinophil, platelet, and hemoglobin results. Trained using the light gradient boosting machine algorithm on clinical data from 1,927 KD cases and 45,274 febrile controls, KDpredictor achieved strong performance metrics (auROC: 95.7%, auPRC: 72.4%, recall: 0.89) on a reserved test set, outperforming previous models by at least 3% in auROC and 39.3% in auPRC. Additional explainable AI analyses revealed that several top predictive features in KDpredictor are consistent with prior clinical findings. We also evaluated KDpredictor on three independent cohorts collected in East Asia (Taiwan and China) during the COVID-19 period. KDpredictor achieves recall values of 90.9%, 83.7%, and 91.7% on KD samples identified in three independent medical centers, respectively, indicating its applicability across independent clinical settings. In summary, KDpredictor demonstrates robust generalizability in KD risk identification across populations by using only standard blood samples independent of clinical symptoms. KDpredictor is freely available at https://cosbi.ee.ncku.edu.tw/KD_under7/.
- 31 May 2024
- Published by Elsevier in Computer Methods and Programs in Biomedicine
- Vol. 254, 108260
BAPCP: A comprehensive and user-friendly web tool for identifying biomarkers from protein microarray technologies
- 31 May 2024
- Published by Elsevier in Computer Methods and Programs in Biomedicine
- Vol. 254, 108260
BAPCP: A comprehensive and user-friendly web tool for identifying biomarkers from protein microarray technologies
- 30 October 2023
- Published by American Chemical Society (ACS) in Journal of Chemical Information and Modeling
- Vol. 64 (7) , 2445-2453
Identifying Human miRNA Target Sites via Learning the Interaction Patterns between miRNA and mRNA Segments
MiRNAs (microRNAs) target specific mRNA (messenger RNA) sites to regulate their translation expression. Although miRNA targeting can rely on seed region base pairing, animal miRNAs, including human miRNAs, typically cooperate with several cofactors, leading to various noncanonical pairing rules. Therefore, identifying the binding sites of animal miRNAs remains challenging. Because experiments for mapping miRNA targets are costly, computational methods are preferred for extracting potential miRNA-mRNA fragment binding pairs first. However, existing prediction tools can have significant false positives due to the prevalent noncanonical miRNA binding behaviors and the information-biased training negative sets that were used while constructing these tools. To overcome these obstacles, we first prepared an information-balanced miRNA binding pair ground-truth data set. A miRNA-mRNA interaction-aware model was then designed to help identify miRNA binding events. On the test set, our model (auROC = 94.4%) outperformed existing models by at least 2.8% in auROC. Furthermore, we showed that this model can suggest potential binding patterns for miRNA-mRNA sequence interacting pairs. Finally, we made the prepared data sets and the designed model available at http://cosbi2.ee.ncku.edu.tw/mirna_binding/download.
- 30 October 2023
- Published by American Chemical Society (ACS) in Journal of Chemical Information and Modeling
- Vol. 64 (7) , 2445-2453
Identifying Human miRNA Target Sites via Learning the Interaction Patterns between miRNA and mRNA Segments
MiRNAs (microRNAs) target specific mRNA (messenger RNA) sites to regulate their translation expression. Although miRNA targeting can rely on seed region base pairing, animal miRNAs, including human miRNAs, typically cooperate with several cofactors, leading to various noncanonical pairing rules. Therefore, identifying the binding sites of animal miRNAs remains challenging. Because experiments for mapping miRNA targets are costly, computational methods are preferred for extracting potential miRNA-mRNA fragment binding pairs first. However, existing prediction tools can have significant false positives due to the prevalent noncanonical miRNA binding behaviors and the information-biased training negative sets that were used while constructing these tools. To overcome these obstacles, we first prepared an information-balanced miRNA binding pair ground-truth data set. A miRNA-mRNA interaction-aware model was then designed to help identify miRNA binding events. On the test set, our model (auROC = 94.4%) outperformed existing models by at least 2.8% in auROC. Furthermore, we showed that this model can suggest potential binding patterns for miRNA-mRNA sequence interacting pairs. Finally, we made the prepared data sets and the designed model available at http://cosbi2.ee.ncku.edu.tw/mirna_binding/download.
- Open Access
- 2 December 2022
- Published by Springer Nature in BMC Plant Biology
- Vol. 22 (1) , 1-14
OrchidBase 5.0: updates of the orchid genome knowledgebase
Containing the largest number of species, the orchid family provides not only materials for studying plant evolution and environmental adaptation, but economically and culturally important ornamental plants for human society. Previously, we collected genome and transcriptome information of Dendrobium catenatum, Phalaenopsis equestris, and Apostasia shenzhenica which belong to two different subfamilies of Orchidaceae, and developed user-friendly tools to explore the orchid genetic sequences in the OrchidBase 4.0. The OrchidBase 4.0 offers the opportunity for plant science community to compare orchid genomes and transcriptomes and retrieve orchid sequences for further study.In the year 2022, two whole-genome sequences of Orchidoideae species, Platanthera zijinensis and Platanthera guangdongensis, were de novo sequenced, assembled and analyzed. In addition, systemic transcriptomes from these two species were also established. Therefore, we included these datasets to develop the new version of OrchidBase 5.0. In addition, three new functions including synteny, gene order, and miRNA information were also developed for orchid genome comparisons and miRNA characterization.OrchidBase 5.0 extended the genetic information to three orchid subfamilies (including five orchid species) and provided new tools for orchid researchers to analyze orchid genomes and transcriptomes. The online resources can be accessed at https://cosbi.ee.ncku.edu.tw/orchidbase5/
- 2 December 2022
- Published by Springer Nature in BMC Plant Biology
- Vol. 22 (1) , 1-14
Open AccessOrchidBase 5.0: updates of the orchid genome knowledgebase
Containing the largest number of species, the orchid family provides not only materials for studying plant evolution and environmental adaptation, but economically and culturally important ornamental plants for human society. Previously, we collected genome and transcriptome information of Dendrobium catenatum, Phalaenopsis equestris, and Apostasia shenzhenica which belong to two different subfamilies of Orchidaceae, and developed user-friendly tools to explore the orchid genetic sequences in the OrchidBase 4.0. The OrchidBase 4.0 offers the opportunity for plant science community to compare orchid genomes and transcriptomes and retrieve orchid sequences for further study.In the year 2022, two whole-genome sequences of Orchidoideae species, Platanthera zijinensis and Platanthera guangdongensis, were de novo sequenced, assembled and analyzed. In addition, systemic transcriptomes from these two species were also established. Therefore, we included these datasets to develop the new version of OrchidBase 5.0. In addition, three new functions including synteny, gene order, and miRNA information were also developed for orchid genome comparisons and miRNA characterization.OrchidBase 5.0 extended the genetic information to three orchid subfamilies (including five orchid species) and provided new tools for orchid researchers to analyze orchid genomes and transcriptomes. The online resources can be accessed at https://cosbi.ee.ncku.edu.tw/orchidbase5/
- 14 November 2022
- Published by Elsevier in Computers in Biology and Medicine
- Vol. 151
YMLA: A comparative platform to carry out functional enrichment analysis for multiple gene lists in yeast
- 14 November 2022
- Published by Elsevier in Computers in Biology and Medicine
- Vol. 151
YMLA: A comparative platform to carry out functional enrichment analysis for multiple gene lists in yeast