Shih-Yi Hsiung

Dr. Shih-Yi Hsiung

GaLSIC, Soka University, Tokyo, Japan
orcid
Postdoc fellow

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 September 2022 in Carbohydrate Research
Verified
Journal Article
Open access
JLJing Li
Jing Li
SHShih-Yi Hsiung
Shih-Yi Hsiung
MKMu-Rong Kao
Mu-Rong Kao
XXXiaohui Xing
Xiaohui Xing
SCShu-Chieh Chang
Shu-Chieh Chang
Yves SY HsiehYves SY Hsieh
Yves SY Hsieh
Polygonatum odoratum is a perennial rhizomatous medicinal plant and different plant parts have been used in the treatment of various ailments. Herein, we have investigated the structural compositions of rhizome, leaf, and stem cell walls. We found 30-44% of polysaccharides in these wall preparations were cyclohexanediaminetetraacetic acid (CDTA) extractable, the proportion of heteromannans (HMs) in the rhizome is nearly three-fold compared to that of the leave and stem. The pectic polysaccharides of the rhizome are also structurally more diverse, with arabinans and type I and type II arabinogalactans being richest as shown by linkage study of the sodium carbonate (Na2CO3) extract. In addition, the 2-linked Araf was rhizome-specific, suggesting the cell walls in the rhizome had adapted to a more complex structure compared to that of the leaf and stem. Water-soluble polysaccharide fractions were also investigated, high proportion of Man as in 4-linked Manp indicated high proportion of HMs. The 21.4 kDa pectic polysaccharides and HMs derived from rhizome cell walls induced specific immune response in mice macrophage cells producing IL-1α and hematopoietic growth factors GM-CSF and G-CSF in vitro.

29 August 2023 in Carbohydrate Polymers
Verified
Journal Article
Open access
SHShih-Yi Hsiung
Shih-Yi Hsiung
SDShun-Xin Deng
Shun-Xin Deng
JLJing Li
Jing Li
SHSheng-Yao Huang
Sheng-Yao Huang
CLChen-Kun Liaw
Chen-Kun Liaw
Yves SY HsiehYves SY Hsieh
Yves SY Hsieh
Machine learning (ML) has been used for many clinical decision-making processes and diagnostic procedures in bioinformatics applications. We examined eight algorithms, including linear discriminant analysis (LDA), logistic regression (LR), k-nearest neighbor (KNN), random forest (RF), gradient boosting machine (GBM), support vector machine (SVM), Naïve Bayes classifier (NB), and artificial neural network (ANN) models, to evaluate their classification and prediction capabilities for four tissue types in Wolfiporia extensa using their monosaccharide composition profiles. All 8 ML-based models were assessed as exemplary models with AUC exceeding 0.8. Five models, namely LDA, KNN, RF, GBM, and ANN, performed excellently in the four-tissue-type classification (AUC > 0.9). Additionally, all eight models were evaluated as good predictive models with AUC value > 0.8 in the three-tissue-type classification. Notably, all 8 ML-based methods outperformed the single linear discriminant analysis (LDA) plotting method. For large sample sizes, the ML-based methods perform better than traditional regression techniques and could potentially increase the accuracy in identifying tissue samples of W. extensa.

17 January 2023 in Planta
Verified
Journal Article
Open access
PDF file
SHShih-Yi Hsiung
Shih-Yi Hsiung
JLJing Li
Jing Li
BIBalazs Imre
Balazs Imre
MKMu-Rong Kao
Mu-Rong Kao
HLHsien-Chun Liao
Hsien-Chun Liao
Yves SY HsiehYves SY Hsieh
Yves SY Hsieh
Main conclusion: The xyloglucans of all aquatic Araceae species examined had unusual structures compared with those of other non-commelinid monocotyledon families previously examined. The aquatic Araceae species Lemna minor was earlier shown to have xyloglucans with a different structure from the fucogalactoxyloglucans of other non-commelinid monocotyledons. We investigated 26 Araceae species (including L. minor), from five of the seven subfamilies. All seven aquatic species examined had xyloglucans that were unusual in having one or two of three features: < 77% XXXG core motif [L. minor (Lemnoideae) and Orontium aquaticum (Orontioideae)]; no fucosylation [L. minor (Lemnoideae), Cryptocoryne aponogetonifolia, and Lagenandra ovata (Aroideae, Rheophytes clade)]; and > 14% oligosaccharide units with S or D side chains [Spirodela polyrhiza and Landoltia punctata (Lemnoideae) and Pistia stratiotes (Aroideae, Dracunculus clade)]. Orontioideae and Lemnoideae are the two most basal subfamilies, with all species being aquatic, and Aroideae is the most derived. Two terrestrial species [Dieffenbachia seguine and Spathicarpa hastifolia (Aroideae, Zantedeschia clade)] also had xyloglucans without fucose indicating this feature was not unique to aquatic species.

12 March 2026
Verified
Book Chapter
Closed access
PDF file
SHShih-Yi Hsiung
Shih-Yi Hsiung
AZAchille Zappa
Achille Zappa
YAYukie Akune-Taylor
Yukie Akune-Taylor
KAKiyoko F. Aoki-Kinoshita
Kiyoko F. Aoki-Kinoshita
Glycoinformatics originated with pioneering initiatives like CarbBank in the 1980s, which represented the first systematic digital repository for carbohydrate structures. Despite this early foundation, the dis-cipline still confronts significant computational and methodological obstacles. Key challenges include modeling complex multivalent binding patterns in protein-carbohydrate and lipid-carbohydrate interac-tions, developing complete glycosylation maps for diagnostic biomarkers, and addressing the inherent variability of non-template-based glycan synthesis pathways. However, recent decades have brought substantial advances in structural analysis capabilities, particularly through artificial intelligence (AI), machine learning (ML), and enhanced glycomics databases. These technological developments offer promising solutions to longstanding problems in the field. This review examines the current landscape of glycoinformatics technologies, including approaches for glycan representation, methods for visualizing glycoconjugates and glycosylation patterns, Semantic Web principles and applications, glycomics data-base systems, and artificial intelligence innovations and implications. The discussion focuses on how these emerging tools are addressing and potentially resolving the persistent challenges that have histori-cally limited progress in computational glycobiology. The goal is to provide essential glycoinformatics concepts, platforms, and application tools to help glycoscientists overcome their research obstacles.

27 July 2026 in Life
Verified
Journal Article
Open access
PDF file
SDShun-Xin Deng
Shun-Xin Deng
LCLih-Geeng Chen
Lih-Geeng Chen
SHShih-Yi Hsiung
Shih-Yi Hsiung
VLVinh Tuyen T. Le
Vinh Tuyen T. Le
CLChia-Jung Lee
Chia-Jung Lee
CWChing-Chiung Wang
Ching-Chiung Wang
Wolfiporia extensa (Peck) Ginns (WE) is a fungus widely used in Traditional Chinese Medicine. Its four medicinal parts, Poria (WP), Rubra Poria (RP), Poriae Cutis (PC), and Poria Cum Pini Radix (PPR), are prescribed for distinct therapeutic purposes; however, their quality standards and pharmacological differences remain unclear. This study established reliable quality markers and compared the neuroprotective activities of four medicinal parts of WE. Seven triterpenoids were quantified using validated high-performance liquid chromatography (HPLC) assays, while monosaccharide profiles were analyzed using 1-phenyl-3-methyl-5-pyrazolone-HPLC. The International Commission on Illumination L*a*b* colorimetry and chemometric models were used for authentication. The neuroprotective effects of aqueous extracts were assessed in SH-SY5Y, BV-2, and HOG cell models. Network pharmacology was used to predict compound–target pathway relationships. The content of poricoic acid A (PAA) was the highest in PC; the content of pachymic acid (PA) was the highest in RP; whereas dehydrotumulosic acid (DTUA) was enriched in WP. WP demonstrated the highest mannose and galactose content. The water extract of WP reduced H2O2-induced cytotoxicity in SH-SY5Y cells, whereas the PC extract alleviated L-α-lysophosphatidylcholine-induced injury in human oligodendroglioma cells. Network analysis identified PA, PAA, and poricoic acid B (PAB) as candidate bioactive compounds involved in neuroprotective effects. The PA/PAB ratios of ≥10, 1–9, and ≤1 were used to identify WP, RP, and PC, respectively. L* values of >80, 55–79, and <55 effectively discriminated WP, RP, and PC, respectively. This integrative study identified chemical and functional quality markers and demonstrated distinct neuroprotective profiles in different parts of WE, supporting their standardized and rational application in traditional medicine and nutraceuticals.

15 April 2020
Verified
Preprint
Open access
SHShih-Yi Hsiung
Shih-Yi Hsiung
LMLuke Mander
Luke Mander
CLCindy Looy
Cindy Looy
PPPrahlada Papper
Prahlada Papper
1Hsiung, Shih-Yi, 2Looy, Cindy,3Mander, Luke, and 4 Papper, Prahlada,1,2 Department of Integrative Biology and Museum of Paleontology, University of California, 3060 Valley Life Sciences Building #3140, Berkeley, California 94720, USA.3 College of Lif