Haipeng Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Haipeng Wang has a total of 59 co-authors. Here are the top co-authors by number of shared publications:
Di Zhang
School of Computer Science and Technology, Shandong University of Technology, Zibo, China
2 shared publications- Fengtang Yang
School of Life Sciences and Medicine, Shandong University of Technology, Zibo 255049, China
2 shared publications Li Kang
State Key Laboratory of Medical Proteomics, National Center for Protein Sciences (Beijing), Research Unit of Proteomics and Research and Development of New Drug of Chinese Academy of Medical Sciences, Beijing Proteome Research Center, Beijing Institute of Lifeomics, Beijing, China
2 shared publications- Libo Jiang
College of Life Sciences and Medicine, Shandong University of Technology, Zibo 255000, China
2 shared publications Lijin Yao
School of Computer Science and Technology, Shandong University of Technology, Zibo, China
2 shared publications- Ping Xu
State Key Laboratory of Proteomics, Beijing Proteome Research Center, National Center for Protein Sciences (Beijing), Research Unit of Proteomics & Research and Development of New Drug of Chinese Academy of Medical Sciences, Beijing Institute of Lifeomics, Beijing 102206, China
2 shared publications
Latest Publications
- Open Access
- 5 March 2026
- Published by Springer Nature in Nature Communications
- Vol. 17 (1) , 2203
DiNovo enables high-coverage and high-confidence de novo peptide sequencing via mirror proteases and deep learning
Despite the recent advancements driven by deep learning, de novo peptide sequencing is still constrained by incomplete peptide fragmentation and insufficient protein digestion in current single protease-based proteomic experiments. Here, we present a software system, named DiNovo, for high-coverage and high-confidence de novo peptide sequencing by leveraging the complementarity of mirror proteases. DiNovo is empowered by several innovative algorithms, including a mirror-spectra recognition algorithm independent of pre-sequencing, two sequencing algorithms based on deep learning and graph theory, respectively, and target-decoy mapping, a method for sequencing result evaluation free of prior peptide identification. Compared with the trypsin protease used alone, DiNovo using two pairs of mirror proteases leads to two to three times high-confidence amino acids sequenced. Compared with previous single-protease de novo sequencing algorithms, DiNovo achieves much higher sequence coverage. DiNovo also shows great potential as a practical and powerful alternative to database search for peptide identification with quality control.
- 5 March 2026
- Published by Springer Nature in Nature Communications
- Vol. 17 (1) , 2203
Open AccessDiNovo enables high-coverage and high-confidence de novo peptide sequencing via mirror proteases and deep learning
Despite the recent advancements driven by deep learning, de novo peptide sequencing is still constrained by incomplete peptide fragmentation and insufficient protein digestion in current single protease-based proteomic experiments. Here, we present a software system, named DiNovo, for high-coverage and high-confidence de novo peptide sequencing by leveraging the complementarity of mirror proteases. DiNovo is empowered by several innovative algorithms, including a mirror-spectra recognition algorithm independent of pre-sequencing, two sequencing algorithms based on deep learning and graph theory, respectively, and target-decoy mapping, a method for sequencing result evaluation free of prior peptide identification. Compared with the trypsin protease used alone, DiNovo using two pairs of mirror proteases leads to two to three times high-confidence amino acids sequenced. Compared with previous single-protease de novo sequencing algorithms, DiNovo achieves much higher sequence coverage. DiNovo also shows great potential as a practical and powerful alternative to database search for peptide identification with quality control.
- Open Access
- 9 June 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (12) , 6482
A Compressed Sequence Tag Index for Fast Peptide Retrieval and Efficient Storage in Protein Identification Search Engines
Proteins regulate various cellular processes and are of great biological interest. The protein search engine is a crucial tool in proteomics research, used to analyze high-throughput tandem mass spectrometry data and to identify protein sequence information. A core step in protein search engines is constructing sequence tag indexes and performing the rapid retrieval of protein databases. However, as the scale of protein sequence data continues to grow, traditional protein search engines face the dual challenges of the high storage cost of sequence tag indexes and low retrieval efficiency. To address these issues, we propose a sequence tag index scheme named STIP, which is based on an inverted index and compression techniques. Based on STIP, we design a peptide retrieval algorithm named STIP-Search. This algorithm utilizes the sequence tag index constructed by STIP for peptide sequence retrieval. STIP uses the greedy algorithm to partition the tag index into blocks; in this way, STIP can generate tag indexes for very large protein databases, such as NCBI-nr. Compared to the current four mainstream tag index generation algorithms used in Open-pFind, MODplus, TIIP and PIPI2, STIP has the lowest storage and time consumption. It utilizes delta encoding, index reduction, and dynamic bit width encoding to compress the tag index, reducing the storage cost by 76.2%. Compared to TIIP, which is currently the algorithm with the lowest time complexity, the time cost of the peptide sequence retrieval of STIP-Search is reduced by 8.94% to 23.31%.
- 9 June 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (12) , 6482
Open AccessA Compressed Sequence Tag Index for Fast Peptide Retrieval and Efficient Storage in Protein Identification Search Engines
Proteins regulate various cellular processes and are of great biological interest. The protein search engine is a crucial tool in proteomics research, used to analyze high-throughput tandem mass spectrometry data and to identify protein sequence information. A core step in protein search engines is constructing sequence tag indexes and performing the rapid retrieval of protein databases. However, as the scale of protein sequence data continues to grow, traditional protein search engines face the dual challenges of the high storage cost of sequence tag indexes and low retrieval efficiency. To address these issues, we propose a sequence tag index scheme named STIP, which is based on an inverted index and compression techniques. Based on STIP, we design a peptide retrieval algorithm named STIP-Search. This algorithm utilizes the sequence tag index constructed by STIP for peptide sequence retrieval. STIP uses the greedy algorithm to partition the tag index into blocks; in this way, STIP can generate tag indexes for very large protein databases, such as NCBI-nr. Compared to the current four mainstream tag index generation algorithms used in Open-pFind, MODplus, TIIP and PIPI2, STIP has the lowest storage and time consumption. It utilizes delta encoding, index reduction, and dynamic bit width encoding to compress the tag index, reducing the storage cost by 76.2%. Compared to TIIP, which is currently the algorithm with the lowest time complexity, the time cost of the peptide sequence retrieval of STIP-Search is reduced by 8.94% to 23.31%.
- Open Access
Identification and Characterization of LINE and SINE Retrotransposons in the African Hedgehog (Atelerix albiventris, Erinaceidae) and Their Association with 3D Genome Organization and Gene Expression
Background: The African hedgehog (Atelerix albiventris) exhibits specialized skin differentiation leading to spine formation, yet its regulatory mechanisms remain unclear. Transposable elements (TEs), particularly LINEs (long interspersed nuclear elements) and SINEs (short interspersed nuclear elements), are known to influence genome organization and gene regulation. Objectives: Given the high proportion of SINEs in the hedgehog genome, this study aims to characterize the distribution, evolutionary dynamics, and potential regulatory roles of LINEs and SINEs, focusing on their associations with chromatin architecture, DNA methylation, and gene expression. Methods: We analyzed LINE and SINE distribution using HiFi sequencing and classified TE families through phylogenetic reconstruction. Hi-C data were used to explore TE interactions with chromatin architecture, while whole-genome 5mCpG methylation was inferred from PacBio HiFi reads of muscle tissue using a deep-learning-based approach. RNA-seq data from skin tissues were analyzed to assess TE expression and potential associations with genes linked to spine development. Results: SINEs form distinct genomic blocks in GC-rich and highly methylated regions, whereas LINEs are enriched in AT-rich, hypomethylated regions. LINEs and SINEs are associated differently with A/B compartments, with SINEs in euchromatin and LINEs in heterochromatin. Methylation analysis suggests that younger TEs tend to have higher methylation levels, and expression analysis indicates that some differentially expressed TEs may be linked to genes involved in epidermal and skeletal development. Conclusions: This study provides a genome-wide perspective on LINE and SINE distribution, methylation patterns, and potential regulatory roles in A. albiventris. While not establishing a direct causal link, the findings suggest that TEs may influence gene expression associated with spine development, offering a basis for future functional studies.
Open AccessIdentification and Characterization of LINE and SINE Retrotransposons in the African Hedgehog (Atelerix albiventris, Erinaceidae) and Their Association with 3D Genome Organization and Gene Expression
Background: The African hedgehog (Atelerix albiventris) exhibits specialized skin differentiation leading to spine formation, yet its regulatory mechanisms remain unclear. Transposable elements (TEs), particularly LINEs (long interspersed nuclear elements) and SINEs (short interspersed nuclear elements), are known to influence genome organization and gene regulation. Objectives: Given the high proportion of SINEs in the hedgehog genome, this study aims to characterize the distribution, evolutionary dynamics, and potential regulatory roles of LINEs and SINEs, focusing on their associations with chromatin architecture, DNA methylation, and gene expression. Methods: We analyzed LINE and SINE distribution using HiFi sequencing and classified TE families through phylogenetic reconstruction. Hi-C data were used to explore TE interactions with chromatin architecture, while whole-genome 5mCpG methylation was inferred from PacBio HiFi reads of muscle tissue using a deep-learning-based approach. RNA-seq data from skin tissues were analyzed to assess TE expression and potential associations with genes linked to spine development. Results: SINEs form distinct genomic blocks in GC-rich and highly methylated regions, whereas LINEs are enriched in AT-rich, hypomethylated regions. LINEs and SINEs are associated differently with A/B compartments, with SINEs in euchromatin and LINEs in heterochromatin. Methylation analysis suggests that younger TEs tend to have higher methylation levels, and expression analysis indicates that some differentially expressed TEs may be linked to genes involved in epidermal and skeletal development. Conclusions: This study provides a genome-wide perspective on LINE and SINE distribution, methylation patterns, and potential regulatory roles in A. albiventris. While not establishing a direct causal link, the findings suggest that TEs may influence gene expression associated with spine development, offering a basis for future functional studies.
- Open Access
DiNovo: high-coverage, high-confidence de novo peptide sequencing using mirror proteases and deep learning
Despite the recent advancements driven by deep learning, de novo peptide sequencing is still constrained by incomplete peptide fragmentation and insufficient protein digestion in current single protease-based proteomic experiments. Here, we present a software system, named DiNovo, for high-coverage and confidence de novo peptide sequencing by leveraging the complementarity of mirror proteases. DiNovo is empowered by several innovative algorithms, including a mirror-spectra recognition algorithm independent of pre-sequencing, two sequencing algorithms based on deep learning and graph theory, respectively, and target-decoy mapping, a method for sequencing result evaluation free of prior peptide identification. Compared with the trypsin protease used alone, DiNovo using two pairs of mirror proteases led to two to three times high-confidence amino acids sequenced. Compared with previous single-protease de novo sequencing algorithms, DiNovo achieved much higher sequence coverages. DiNovo also showed great potential as a powerful complement or alternative to database search for peptide identification with quality control.
Open AccessDiNovo: high-coverage, high-confidence de novo peptide sequencing using mirror proteases and deep learning
Despite the recent advancements driven by deep learning, de novo peptide sequencing is still constrained by incomplete peptide fragmentation and insufficient protein digestion in current single protease-based proteomic experiments. Here, we present a software system, named DiNovo, for high-coverage and confidence de novo peptide sequencing by leveraging the complementarity of mirror proteases. DiNovo is empowered by several innovative algorithms, including a mirror-spectra recognition algorithm independent of pre-sequencing, two sequencing algorithms based on deep learning and graph theory, respectively, and target-decoy mapping, a method for sequencing result evaluation free of prior peptide identification. Compared with the trypsin protease used alone, DiNovo using two pairs of mirror proteases led to two to three times high-confidence amino acids sequenced. Compared with previous single-protease de novo sequencing algorithms, DiNovo achieved much higher sequence coverages. DiNovo also showed great potential as a powerful complement or alternative to database search for peptide identification with quality control.
- Open Access
- 21 June 2024
- Published by Oxford University Press (OUP) in Genomics, Proteomics & Bioinformatics
- Vol. 22 (4)
The Bioinformatic Applications of Hi-C and Linked Reads
Long-range sequencing grants insight into additional genetic information beyond what can be accessed by both short reads and modern long-read technology. Several new sequencing technologies, such as “Hi-C” and “Linked Reads”, produce long-range datasets for high-throughput and high-resolution genome analyses, which are rapidly advancing the field of genome assembly, genome scaffolding, and more comprehensive variant identification. In this review, we focused on five major long-range sequencing technologies: high-throughput chromosome conformation capture (Hi-C), 10X Genomics Linked Reads, haplotagging, transposase enzyme linked long-read sequencing (TELL-seq), and single- tube long fragment read (stLFR). We detailed the mechanisms and data products of the five platforms and their important applications, evaluated the quality of sequencing data from different platforms, and discussed the currently available bioinformatics tools. This work will benefit the selection of appropriate long-range technology for specific biological studies.
- 21 June 2024
- Published by Oxford University Press (OUP) in Genomics, Proteomics & Bioinformatics
- Vol. 22 (4)
Open AccessThe Bioinformatic Applications of Hi-C and Linked Reads
Long-range sequencing grants insight into additional genetic information beyond what can be accessed by both short reads and modern long-read technology. Several new sequencing technologies, such as “Hi-C” and “Linked Reads”, produce long-range datasets for high-throughput and high-resolution genome analyses, which are rapidly advancing the field of genome assembly, genome scaffolding, and more comprehensive variant identification. In this review, we focused on five major long-range sequencing technologies: high-throughput chromosome conformation capture (Hi-C), 10X Genomics Linked Reads, haplotagging, transposase enzyme linked long-read sequencing (TELL-seq), and single- tube long fragment read (stLFR). We detailed the mechanisms and data products of the five platforms and their important applications, evaluated the quality of sequencing data from different platforms, and discussed the currently available bioinformatics tools. This work will benefit the selection of appropriate long-range technology for specific biological studies.
- Open Access
- 5 April 2023
- Published by MDPI AG in Applied Sciences
- Vol. 13 (7) , 4604
Denovo-GCN: De Novo Peptide Sequencing by Graph Convolutional Neural Networks
The de novo peptide-sequencing method can be used to directly infer the peptide sequence from a tandem mass spectrum. It has the advantage of not relying on protein databases and plays a key role in the determination of the protein sequences of unknown species, monoclonal antibodies, and cancer neoantigens. In this paper, we propose a method based on graph convolutional neural networks and convolutional neural networks, Denovo-GCN, for de novo peptide sequencing. We constructed an undirected graph based on the mass difference between the spectral peaks in a tandem mass spectrum. The features of the nodes on the spectrum graph, which represent the spectral peaks, were the matching information of the peptide sequence and the mass spectrum. Next, the Denovo-GCN used CNN to extract the features of the nodes. The correlation between the nodes was represented by an adjacency matrix, which aggregated the features of neighboring nodes. Denovo-GCN provides a complete end-to-end training and prediction framework to sequence patterns of peptides. Our experiments on various data sets from different species show that Denovo-GCN outperforms DeepNovo with a relative improvement of 13.7–25.5% in terms of the peptide-level recall.
- 5 April 2023
- Published by MDPI AG in Applied Sciences
- Vol. 13 (7) , 4604
Open AccessDenovo-GCN: De Novo Peptide Sequencing by Graph Convolutional Neural Networks
The de novo peptide-sequencing method can be used to directly infer the peptide sequence from a tandem mass spectrum. It has the advantage of not relying on protein databases and plays a key role in the determination of the protein sequences of unknown species, monoclonal antibodies, and cancer neoantigens. In this paper, we propose a method based on graph convolutional neural networks and convolutional neural networks, Denovo-GCN, for de novo peptide sequencing. We constructed an undirected graph based on the mass difference between the spectral peaks in a tandem mass spectrum. The features of the nodes on the spectrum graph, which represent the spectral peaks, were the matching information of the peptide sequence and the mass spectrum. Next, the Denovo-GCN used CNN to extract the features of the nodes. The correlation between the nodes was represented by an adjacency matrix, which aggregated the features of neighboring nodes. Denovo-GCN provides a complete end-to-end training and prediction framework to sequence patterns of peptides. Our experiments on various data sets from different species show that Denovo-GCN outperforms DeepNovo with a relative improvement of 13.7–25.5% in terms of the peptide-level recall.
- Open Access
- 17 December 2018
- Published by Springer Nature in Nature Communications
- Vol. 9 (1) , 1-13
Microbiome characterization by high-throughput transfer RNA sequencing and modification analysis
Advances in high-throughput sequencing have facilitated remarkable insights into the diversity and functioning of naturally occurring microbes; however, current sequencing strategies are insufficient to reveal physiological states of microbial communities associated with protein translation dynamics. Transfer RNAs (tRNAs) are core components of protein synthesis machinery, present in all living cells, and are phylogenetically tractable, which make them ideal targets to gain physiological insights into environmental microbes. Here we report a direct sequencing approach, tRNA-seq, and a software suite, tRNA-seq-tools, to recover sequences, abundance profiles, and post-transcriptional modifications of microbial tRNA transcripts. Our analysis of cecal samples using tRNA-seq distinguishes high-fat- and low-fat-fed mice in a comparable fashion to 16S ribosomal RNA gene amplicons, and reveals taxon- and diet-dependent variations in tRNA modifications. Our results provide taxon-specific in situ insights into the dynamics of tRNA gene expression and post-transcriptional modifications within complex environmental microbiomes.
- 17 December 2018
- Published by Springer Nature in Nature Communications
- Vol. 9 (1) , 1-13
Open AccessMicrobiome characterization by high-throughput transfer RNA sequencing and modification analysis
Advances in high-throughput sequencing have facilitated remarkable insights into the diversity and functioning of naturally occurring microbes; however, current sequencing strategies are insufficient to reveal physiological states of microbial communities associated with protein translation dynamics. Transfer RNAs (tRNAs) are core components of protein synthesis machinery, present in all living cells, and are phylogenetically tractable, which make them ideal targets to gain physiological insights into environmental microbes. Here we report a direct sequencing approach, tRNA-seq, and a software suite, tRNA-seq-tools, to recover sequences, abundance profiles, and post-transcriptional modifications of microbial tRNA transcripts. Our analysis of cecal samples using tRNA-seq distinguishes high-fat- and low-fat-fed mice in a comparable fashion to 16S ribosomal RNA gene amplicons, and reveals taxon- and diet-dependent variations in tRNA modifications. Our results provide taxon-specific in situ insights into the dynamics of tRNA gene expression and post-transcriptional modifications within complex environmental microbiomes.
- Open Access
- 17 June 2015
- Published by Optica Publishing Group in Optics Express
- Vol. 23 (13) , 16723-16729
Direct evidence on the energy transfer of near-infrared emission in PbS quantum dot-doped glass
PbS quantum dot (QD)-doped glass was prepared by the heat treatment of as-prepared glass, which was confirmed by transmission electron microscope (TEM), absorption and photoluminescence (PL) spectra. Choosing the glass heat treatment at 600°C for 24h as a representative sample, steady-state and transient-state PL, lifetime and time-resolve emission spectra (TRES) of PbS QD-doped glass were studied in detail. The results indicated that the lifetime spectra showed a similar variation tendency with the PL spectra. The steady-state, transient-state PL and TRES results first revealed the energy transfer process from smaller QDs with higher energy to bigger QDs with lower energy in the PbS QD-doped glass.
- 17 June 2015
- Published by Optica Publishing Group in Optics Express
- Vol. 23 (13) , 16723-16729
Open AccessDirect evidence on the energy transfer of near-infrared emission in PbS quantum dot-doped glass
PbS quantum dot (QD)-doped glass was prepared by the heat treatment of as-prepared glass, which was confirmed by transmission electron microscope (TEM), absorption and photoluminescence (PL) spectra. Choosing the glass heat treatment at 600°C for 24h as a representative sample, steady-state and transient-state PL, lifetime and time-resolve emission spectra (TRES) of PbS QD-doped glass were studied in detail. The results indicated that the lifetime spectra showed a similar variation tendency with the PL spectra. The steady-state, transient-state PL and TRES results first revealed the energy transfer process from smaller QDs with higher energy to bigger QDs with lower energy in the PbS QD-doped glass.