Jing Xian Teo

Jing Xian Teo

Auto-generated profile, based on publicly available information

Jing Xian Teo

SingHealth Duke-NUS Institute of Precision Medicine, Singapore

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

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

Representative Publications

View All Publications
29 July 2022 in Journal of Medical Internet Research
Verified
Journal Article
Open access
WZWeizhuang Zhou
Weizhuang Zhou
YCYu En Chan
Yu En Chan
CFChuan Sheng Foo
Chuan Sheng Foo
JZJingxian Zhang
Jingxian Zhang
JTJing Xian Teo
Jing Xian Teo
PKPavitra Krishnaswamy
Pavitra Krishnaswamy
Background: Consumer-grade wearable devices enable detailed recordings of heart rate and step counts in free-living conditions. Recent studies have shown that summary statistics from these wearable recordings have potential uses for longitudinal monitoring of health and disease states. However, the relationship between higher resolution physiological dynamics from wearables and known markers of health and disease remains largely uncharacterized. Objective: We aimed to derive high-resolution digital phenotypes from observational wearable recordings and to examine their associations with modifiable and inherent markers of cardiometabolic disease risk. Methods: We introduced a principled framework to extract interpretable high-resolution phenotypes from wearable data recorded in free-living conditions. The proposed framework standardizes the handling of data irregularities; encodes contextual information regarding the underlying physiological state at any given time; and generates a set of 66 minimally redundant features across active, sedentary, and sleep states. We applied our approach to a multimodal data set, from the SingHEART study (NCT02791152), which comprises heart rate and step count time series from wearables, clinical screening profiles, and whole genome sequences from 692 healthy volunteers. We used machine learning to model nonlinear relationships between the high-resolution phenotypes on the one hand and clinical or genomic risk markers for blood pressure, lipid, weight and sugar abnormalities on the other. For each risk type, we performed model comparisons based on Brier scores to assess the predictive value of high-resolution features over and beyond typical baselines. We also qualitatively characterized the wearable phenotypes for participants who had actualized clinical events. Results: We found that the high-resolution features have higher predictive value than typical baselines for clinical markers of cardiometabolic disease risk: the best models based on high-resolution features had 17.9% and 7.36% improvement in Brier score over baselines based on age and gender and resting heart rate, respectively (P<.001 in each case). Furthermore, heart rate dynamics from different activity states contain distinct information (maximum absolute correlation coefficient of 0.15). Heart rate dynamics in sedentary states are most predictive of lipid abnormalities and obesity, whereas patterns in active states are most predictive of blood pressure abnormalities (P<.001). Moreover, in comparison with standard measures, higher resolution patterns in wearable heart rate recordings are better able to represent subtle physiological dynamics related to genomic risk for cardiometabolic disease (improvement of 11.9%-22.0% in Brier scores; P<.001). Finally, illustrative case studies reveal connections between these high-resolution phenotypes and actualized clinical events, even for borderline profiles lacking apparent cardiometabolic risk markers. Conclusions: High-resolution digital phenotypes recorded by consumer wearables in free-living states have the potential to enhance the prediction of cardiometabolic disease risk and could enable more proactive and personalized health management. Trial Registration:

22 February 2021 in JNCI Cancer Spectrum
Verified
Journal Article
Open access
PDF file
SCSock Hoai Chan
Sock Hoai Chan
YNYing Ni
Ying Ni
SLShao-Tzu Li
Shao-Tzu Li
JTJing Xian Teo
Jing Xian Teo
NINur Diana Binte Ishak
Nur Diana Binte Ishak
JNJoanne Ngeow
Joanne Ngeow
Fanconi anemia (FA) is a rare genetic disorder associated with hematological disorders and solid tumor predisposition. Owing to phenotypic heterogeneity, some patients remain undetected until adulthood, usually following cancer diagnoses. The uneven prevalence of FA cases with different underlying FA gene mutations worldwide suggests variable genetic distribution across populations. Here, we aim to assess the genetic spectrum of FA-associated genes across populations of varying ancestries and explore potential genotype–phenotype associations in cancer. Carrier frequency and variant spectrum of potentially pathogenic germline variants in 17 FA genes (excluding BRCA1/FANCS, BRCA2/FANCD1, BRIP1/FANCJ, PALB2/FANCN, RAD51C/FANCO) were evaluated in 3523 Singaporeans and 7 populations encompassing Asian, European, African, and admixed ancestries from the Genome Aggregation Database. Germline and somatic variants of 17 FA genes in 7 cancer cohorts from The Cancer Genome Atlas were assessed to explore genotype–phenotype associations. Germline variants in FANCA were consistently more frequent in all populations. Similar trends in carrier frequency and variant spectrum were detected in Singaporeans and East Asians, both distinct from other ancestry groups, particularly in the lack of recurrent variants. Our exploration of The Cancer Genome Atlas dataset suggested higher germline and somatic mutation burden between FANCA and FANCC with head and neck and lung squamous cell carcinomas as well as FANCI and SLX4/FANCP with uterine cancer, but the analysis was insufficiently powered to detect any statistical significance. Our findings highlight the diverse genetic spectrum of FA-associated genes across populations of varying ancestries, emphasizing the need to include all known FA-related genes for accurate molecular diagnosis of FA.

1 December 2018 in Genetics in Medicine
Verified
Journal Article
Closed access
PDF file
YBYasmin Bylstra
Yasmin Bylstra
JKJyn Ling Kuan
Jyn Ling Kuan
WLWeng Khong Lim
Weng Khong Lim
JBJaydutt Digambar Bhalshankar
Jaydutt Digambar Bhalshankar
JTJing Xian Teo
Jing Xian Teo
SJSaumya Shekhar Jamuar
Saumya Shekhar Jamuar

30 January 2020 in bioRxiv
Verified
Preprint
Open access
PDF file
YBYasmin Bylstra
Yasmin Bylstra
WLWeng Khong Lim
Weng Khong Lim
SKSylvia Kam
Sylvia Kam
KTKoei Wan Tham
Koei Wan Tham
RWR. Ryanne Wu
R. Ryanne Wu
PTPatrick Tan
Patrick Tan
Background: Family history has traditionally been an essential part of clinical care to assess health risks. However, declining sequencing costs have precipitated a shift towards genomics-first approaches in population screening programs, with less emphasis on family history assessment. We evaluated the utility of family history for genomic sequencing selection. Methods: We analysed whole genome sequences of 1750 healthy research participants, with and without preselection based on standardised family history collection, screening 95 cancer genes. Results: The frequency of likely pathogenic/ pathogenic (LP/P) variants in 884 participants with no family history available (FH not available group) (2%) versus 866 participants with family history available (FH available group) (3.1%) was not significant ( p =0.158). However, within the FH available group, amongst 73 participants with an increased family history cancer risk (increased FH risk), 1 in 7 participants carried a LP/P variant inferring a six-fold increase compared with 1 in 47 participants assessed at average family history cancer risk (average FH risk) and a seven-fold increase compared to the FH not available group. The enrichment was further pronounced (up to 18-fold) when assessing the 25 cancer genes in the ACMG 59-gene panel. Furthermore, 63 participants had an increased family history cancer risk in absence of an apparent LP/P variant. Conclusion: Our findings show that systematic family history collection remains critical for health risk assessment, providing important actionable data and augmenting the yield from genomic data. Family history also highlights the potential impact of additional hereditary, environmental and behavioural influences not reflected by genomic sequencing.

1 November 2021 in medRxiv
Verified
Preprint
Open access
PDF file
WZWeizhuang Zhou
Weizhuang Zhou
YCYu En Chan
Yu En Chan
CFChuan Sheng Foo
Chuan Sheng Foo
JZJingxian Zhang
Jingxian Zhang
JTJing Xian Teo
Jing Xian Teo
PKPavitra Krishnaswamy
Pavitra Krishnaswamy
Background: Consumer-grade wearable devices enable detailed recordings of heart rate and step counts in free-living conditions. Recent studies have shown that summary statistics from these wearable recordings have potential uses for longitudinal monitoring of health and disease states. However, the relationship between higher resolution physiological dynamics from wearables and known markers of health and disease remains largely uncharacterized. Objective: We aimed to (i) derive high resolution digital phenotypes from observational wearable recordings, (ii) characterize their ability to predict modifiable markers of cardiometabolic disease, and (iii) study their connections with genetic predispositions for cardiometabolic disease and with lifestyle factors. Methods: We introduce a principled framework to extract interpretable high resolution phenotypes from wearable data recorded in free-living conditions. The proposed framework standardizes handling of data irregularities, encodes contextual information about underlying physiological state at any given time, and generates a set of 66 minimally redundant features across active, sedentary and sleep states. We applied our approach on a multimodal dataset, from the SingHEART study ( NCT02791152 ), that comprises of heart rate and step count time series from wearables, clinical screening profiles, whole genome sequences and lifestyle survey responses from 692 healthy volunteers. We employed machine learning to model non-linear relationships between the high resolution phenotypes and clinical risk markers for blood pressure, lipid and weight abnormalities. For each risk type, we performed model comparisons based on Brier Skill Scores (BSS) to assess predictive value of the high resolution features over and beyond typical baselines. We then examined associations between the wearable-derived features, polygenic risk for cardiometabolic disease, and lifestyle habits and health perceptions. Results: Compared to typical summary statistic measures like resting heart rate, we find that the high-resolution features collectively have greater predictive value for modifiable clinical markers associated with cardiometabolic disease risk (average improvement in Brier Skill Score=52.3%, P <.001). Further, we show that heart rate dynamics from different activity states contain distinct information about type of cardiometabolic risk, with dynamics in sedentary states being most predictive of lipid abnormalities and patterns in active states being most predictive of blood pressure abnormalities ( P <.001). Finally, our results reveal that subtle heart rate dynamics in wearable recordings serve as physiological correlates of genetic predisposition for cardiometabolic disease, lifestyle habits and health perceptions. Conclusions: High resolution digital phenotypes recorded by consumer wearables in free-living states have the potential to enhance prediction of cardiometabolic disease risk, and could enable more proactive and personalized health management. Trial Registration: ClinicalTrials.gov NCT02791152 ; https://clinicaltrials.gov/ct2/show/NCT02791152