MyMedR

Displaying publications 1 - 20 of 82 in total

Abstract:

Sort:

Evidence for the Higgs Boson Decay to a Z Boson and a Photon at the LHC

Aad G, Abbott B, Abeling K, Abicht NJ, Abidi SH, Aboulhorma A, et al.

Phys Rev Lett, 2024 Jan 12;132(2):021803.
PMID: 38277607 DOI: 10.1103/PhysRevLett.132.021803

The first evidence for the Higgs boson decay to a Z boson and a photon is presented, with a statistical significance of 3.4 standard deviations. The result is derived from a combined analysis of the searches performed by the ATLAS and CMS Collaborations with proton-proton collision datasets collected at the CERN Large Hadron Collider (LHC) from 2015 to 2018. These correspond to integrated luminosities of around 140 fb^{-1} for each experiment, at a center-of-mass energy of 13 TeV. The measured signal yield is 2.2±0.7 times the standard model prediction, and agrees with the theoretical expectation within 1.9 standard deviations.
microRNAs expression profiles in Schistosoma japonicum of different sex 14 and 28 days post-infection

Ai L, Hu W, Zhang RL, Huang DN, Chen SH, Xu B, et al.

Trop Biomed, 2020 Dec 01;37(4):947-962.
PMID: 33612748 DOI: 10.47665/tb.37.4.947

Different miRNAs are involved in the life cycles of Schistosoma japonicum. The aim of this study was to examine the expression profile of miRNAs in individual S. japonicum of different sex before and after pairing (18 and 24 dpi). The majority of differential expressed miRNAs were highly abundant at 14 dpi, except for sja-miR-125b and sja-miR-3505, in both male and female. Moreover, it was estimated that sja-miR-125b and sja-miR-3505 might be related to laying eggs. sja-miR-2a-5p and sja-miR-3484-5p were expressed at 14 dpi in males and were significantly clustered in DNA topoisomerase III, Rap guanine nucleotide exchange factor 1 and L-serine/L-threonine ammonia-lyase. Target genes of sja-miR-2d-5p, sja-miR-31- 5p and sja-miR-125a, which were expressed at 14 dpi in males but particularly females, were clustered in kelch-like protein 12, fructose-bisphosphate aldolase, class I, and heat shock protein 90 kDa beta. Predicted target genes of sja-miR-3483-3p (expressed at 28 dpi in females but not in males) were clustered in 26S proteasome regulatory subunit N1, ATPdependent RNA helicase DDX17. Predicted target genes of sja-miR-219-5p, which were differentially expressed at 28 dpi in females but particularly males, were clustered in DNA excision repair protein ERCC-6, protein phosphatase 1D, and ATPase family AAA domaincontaining protein 3A/B. Moreover, at 28 dpi, eight miRNAs were significantly up-regulated in females compared to males. The predicted target genes of these miRNAs were significantly clustered in heat shock protein 90 kDa beta, 26S proteasome regulatory subunit N1, and protein arginine N-methyltransferase 1. To sum up, differentially expressed miRNAs may have an essential role and provide necessary information on clarifying this trematode's growth, development, maturation, and infection ability to mammalian hosts in its complex life cycle, and may be helpful for developing new drug targets and vaccine candidates for schistosomiasis.
Fulltext Regulation of hepatic insulin signaling and glucose homeostasis by sphingosine kinase 2

Aji G, Huang Y, Ng ML, Wang W, Lan T, Li M, et al.

Proc Natl Acad Sci U S A, 2020 09 29;117(39):24434-24442.
PMID: 32917816 DOI: 10.1073/pnas.2007856117

Sphingolipid dysregulation is often associated with insulin resistance, while the enzymes controlling sphingolipid metabolism are emerging as therapeutic targets for improving insulin sensitivity. We report herein that sphingosine kinase 2 (SphK2), a key enzyme in sphingolipid catabolism, plays a critical role in the regulation of hepatic insulin signaling and glucose homeostasis both in vitro and in vivo. Hepatocyte-specific Sphk2 knockout mice exhibit pronounced insulin resistance and glucose intolerance. Likewise, SphK2-deficient hepatocytes are resistant to insulin-induced activation of the phosphoinositide 3-kinase (PI3K)-Akt-FoxO1 pathway and elevated hepatic glucose production. Mechanistically, SphK2 deficiency leads to the accumulation of sphingosine that, in turn, suppresses hepatic insulin signaling by inhibiting PI3K activation in hepatocytes. Either reexpressing functional SphK2 or pharmacologically inhibiting sphingosine production restores insulin sensitivity in SphK2-deficient hepatocytes. In conclusion, the current study provides both experimental findings and mechanistic data showing that SphK2 and sphingosine in the liver are critical regulators of insulin sensitivity and glucose homeostasis.
Fulltext Multi-perspectives systematic review on the applications of sentiment analysis for vaccine hesitancy

Alamoodi AH, Zaidan BB, Al-Masawa M, Taresh SM, Noman S, Ahmaro IYY, et al.

Comput Biol Med, 2021 Dec;139:104957.
PMID: 34735945 DOI: 10.1016/j.compbiomed.2021.104957

A substantial impediment to widespread Coronavirus disease (COVID-19) vaccination is vaccine hesitancy. Many researchers across scientific disciplines have presented countless studies in favor of COVID-19 vaccination, but misinformation on social media could hinder vaccination efforts and increase vaccine hesitancy. Nevertheless, studying people's perceptions on social media to understand their sentiment presents a powerful medium for researchers to identify the causes of vaccine hesitancy and therefore develop appropriate public health messages and interventions. To the best of the authors' knowledge, previous studies have presented vaccine hesitancy in specific cases or within one scientific discipline (i.e., social, medical, and technological). No previous study has presented findings via sentiment analysis for multiple scientific disciplines as follows: (1) social, (2) medical, public health, and (3) technology sciences. Therefore, this research aimed to review and analyze articles related to different vaccine hesitancy cases in the last 11 years and understand the application of sentiment analysis on the most important literature findings. Articles were systematically searched in Web of Science, Scopus, PubMed, IEEEXplore, ScienceDirect, and Ovid from January 1, 2010, to July 2021. A total of 30 articles were selected on the basis of inclusion and exclusion criteria. These articles were formed into a taxonomy of literature, along with challenges, motivations, and recommendations for social, medical, and public health and technology sciences. Significant patterns were identified, and opportunities were promoted towards the understanding of this phenomenon.
Measurement of WZ and ZZ production in pp collisions at [Formula: see text] in final states with b-tagged jets

CMS Collaboration, Chatrchyan S, Khachatryan V, Sirunyan AM, Tumasyan A, Adam W, et al.

Eur Phys J C Part Fields, 2014 08 07;74(8):2973.
PMID: 25814904

Measurements are reported of the WZ and ZZ production cross sections in proton-proton collisions at [Formula: see text][Formula: see text] in final states where one Z boson decays to b-tagged jets. The other gauge boson, either W or Z, is detected through its leptonic decay (either [Formula: see text], [Formula: see text] or [Formula: see text], [Formula: see text], or [Formula: see text]). The results are based on data corresponding to an integrated luminosity of 18.9 fb[Formula: see text] collected with the CMS detector at the Large Hadron Collider. The measured cross sections, [Formula: see text] and [Formula: see text], are consistent with next-to-leading order quantum chromodynamics calculations.
Measurement of jet multiplicity distributions in [Formula: see text] production in pp collisions at [Formula: see text]

CMS Collaboration, Chatrchyan S, Khachatryan V, Sirunyan AM, Tumasyan A, Adam W, et al.

Eur Phys J C Part Fields, 2014 08 20;74(8):3014.
PMID: 25814909

The normalised differential top quark-antiquark production cross section is measured as a function of the jet multiplicity in proton-proton collisions at a centre-of-mass energy of 7[Formula: see text] at the LHC with the CMS detector. The measurement is performed in both the dilepton and lepton+jets decay channels using data corresponding to an integrated luminosity of 5.0[Formula: see text]. Using a procedure to associate jets to decay products of the top quarks, the differential cross section of the [Formula: see text] production is determined as a function of the additional jet multiplicity in the lepton+jets channel. Furthermore, the fraction of events with no additional jets is measured in the dilepton channel, as a function of the threshold on the jet transverse momentum. The measurements are compared with predictions from perturbative quantum chromodynamics and no significant deviations are observed.
Measurement of differential cross sections for the production of a pair of isolated photons in pp collisions at [Formula: see text]

CMS Collaboration, Chatrchyan S, Khachatryan V, Sirunyan AM, Tumasyan A, Adam W, et al.

Eur Phys J C Part Fields, 2014 11 12;74(11):3129.
PMID: 25814874

A measurement of differential cross sections for the production of a pair of isolated photons in proton-proton collisions at [Formula: see text] is presented. The data sample corresponds to an integrated luminosity of 5.0[Formula: see text] collected with the CMS detector. A data-driven isolation template method is used to extract the prompt diphoton yield. The measured cross section for two isolated photons, with transverse energy above 40 and 25[Formula: see text] respectively, in the pseudorapidity range [Formula: see text], [Formula: see text] and with an angular separation [Formula: see text], is [Formula: see text][Formula: see text]. Differential cross sections are measured as a function of the diphoton invariant mass, the diphoton transverse momentum, the azimuthal angle difference between the two photons, and the cosine of the polar angle in the Collins-Soper reference frame of the diphoton system. The results are compared to theoretical predictions at leading, next-to-leading, and next-to-next-to-leading order in quantum chromodynamics.
Fulltext Description of apps targeting stroke patients: A review of apps store

Cao W, Kadir AA, Wang Y, Wang J, Dai B, Zheng Y, et al.

Digit Health, 2023;9:20552076231181473.
PMID: 37342095 DOI: 10.1177/20552076231181473

BACKGROUND: As a principal cause of mortality and disability worldwide, stroke imposes considerable burdens on society and effects on the lives of patients, families, and communities. Owing to their growing global popularity, health-related applications (apps) offer a promising approach to stroke management but show a knowledge gap regarding mobile apps for stroke survivors.
METHODS: This review was conducted across the Android and iOS app stores in September-December 2022 to identify and describe all apps targeting stroke survivors. Apps were included if they were designed for stroke management and contained at least one of the following components: medication taking, risk management, blood pressure management, and stroke rehabilitation. Apps were excluded if they were unrelated to health, not in Chinese or English, or the targeted users were healthcare professionals. The included apps were downloaded, and their functionalities were investigated.
RESULTS: The initial search yielded 402 apps, with 115 eligible after title and description screening. Some apps were later excluded due to duplicates, registration problems, or installation failures. In total, 83 apps were included for full review and evaluated by three independent reviewers. Educational information was the most common function (36.1%), followed by rehabilitation guidance (34.9%), communication with healthcare providers (HCPs), and others (28.9%). The majority of these apps (50.6%) had only one functionality. A minority had contributions from an HCP or patients.
CONCLUSION: With the widespread accessibility and availability of smartphone apps across the mHealth landscape, an increasing number of apps targeting stroke survivors are being released. One of the most important findings is that the majority of the apps were not specifically geared toward older adults. Many of the currently available apps lack healthcare professionals' and patients' involvement in their development, and most offer limited functionality, thus requiring further attention to the development of customized apps.
Search for top squark and Higgsino production using diphoton Higgs boson decays

Chatrchyan S, Khachatryan V, Sirunyan AM, Tumasyan A, Adam W, Bergauer T, et al.

Phys Rev Lett, 2014 Apr 25;112(16):161802.
PMID: 24815637

Results are presented of a search for a "natural" supersymmetry scenario with gauge mediated symmetry breaking. It is assumed that only the supersymmetric partners of the top quark (the top squark) and the Higgs boson (Higgsino) are accessible. Events are examined in which there are two photons forming a Higgs boson candidate, and at least two b-quark jets. In 19.7 fb-1 of proton-proton collision data at s=8 TeV, recorded in the CMS experiment, no evidence of a signal is found and lower limits at the 95% confidence level are set, excluding the top squark mass below 360 to 410 GeV, depending on the Higgsino mass.
Search for invisible decays of Higgs bosons in the vector boson fusion and associated ZH production modes

Chatrchyan S, Khachatryan V, Sirunyan AM, Tumasyan A, Adam W, Bergauer T, et al.

Eur Phys J C Part Fields, 2014;74(8):2980.
PMID: 25814906

A search for invisible decays of Higgs bosons is performed using the vector boson fusion and associated ZH production modes. In the ZH mode, the Z boson is required to decay to a pair of charged leptons or a [Formula: see text] quark pair. The searches use the 8┬á[Formula: see text] pp collision dataset collected by the CMS detector at the LHC, corresponding to an integrated luminosity of up to 19.7┬á[Formula: see text]. Certain channels include data from 7┬á[Formula: see text] collisions corresponding to an integrated luminosity of 4.9┬á[Formula: see text]. The searches are sensitive to non-standard-model invisible decays of the recently observed Higgs boson, as well as additional Higgs bosons with similar production modes and large invisible branching fractions. In all channels, the observed data are consistent with the expected standard model backgrounds. Limits are set on the production cross section times invisible branching fraction, as a function of the Higgs boson mass, for the vector boson fusion and ZH production modes. By combining all channels, and assuming standard model Higgs boson cross sections and acceptances, the observed (expected) upper limit on the invisible branching fraction at [Formula: see text]┬á[Formula: see text] is found to be 0.58┬á(0.44) at 95┬á% confidence level. We interpret this limit in terms of a Higgs-portal model of dark matter interactions.
Fulltext Distribution of axial length, anterior chamber depth, and corneal curvature in an aged population in South China

Chen H, Lin H, Lin Z, Chen J, Chen W

BMC Ophthalmol, 2016 May 01;16(1):47.
PMID: 27138378 DOI: 10.1186/s12886-016-0221-5

BACKGROUND: Ocular biometry is important for preoperative assessment in cataract and anterior segment surgery. The purpose of this study was to investigate normative ocular biometric parameters and their associations in an older Chinese population.
METHODS: This was a cross-sectional observational study. From 2013 to 2014, we recruited inhabitants aged 50 years or older in Guangzhou, China. Among 1,117 participants in the study, data from 1,015 phakic right eyes were used for analyses. Ocular parameters including axial length (AL), anterior chamber depth (ACD), and corneal curvature (K) were measured using an IOL Master.
RESULTS: The mean AL, ACD, and K were 23.48 mm [95 % confidence interval (CI), 23.40-23.55], 3.03 mm (CI, 3.01-3.05), and 44.20 mm (CI, 44.11-44.29), respectively. A mean reduction in ACD with age was observed (P = 0.002) in male subjects but not in female subjects (P = 0.558). Male subjects had significantly longer ALs (23.68 mm versus 23.23 mm, P
Fulltext In vitro anti-inflammatory activity of fractionated Euphorbia hirta aqueous extract on rabbit synovial fibroblasts

Chen J, Er HM, Mohamed SM, Chen YS

Biomed J, 2015 Jul-Aug;38(4):301-6.
PMID: 25673170 DOI: 10.4103/2319-4170.151031

BACKGROUND: Euphorbia hirta has been reported to possess anti-inflammatory activity. This study was carried out to determine the prostaglandin E 2 (PGE 2 ) inhibition activity of the fractions of the E. hirta aqueous extract on rabbit synovial fibroblast cells (HIG-82).
METHODS: E. hirta aqueous extract was fractionated into five fractions (fractions A, B, C, D, and E) by reversed phase flash chromatography. Rabbit synovial fibroblast cells (HIG-82) were activated with phorbol myristate acetate and treated with the fractions. The amount of PGE 2 released into the medium was measured by enzyme-linked immunosorbent assay.
RESULTS: Fraction A (0.1, 1, and 10 μg/ml) had the greatest PGE 2 inhibitory effect among the five fractions, and showed a greater extent of PGE 2 inhibition compared to the aqueous extract. In contrast, Fraction E had the greatest stimulatory effect on PGE 2 release.
CONCLUSIONS: Fraction A of the aqueous extract inhibited the production of PGE 2 from activated HIG-82 cells to a greater extent than the crude aqueous extract. Bioactive compounds with anti-inflammatory activity are likely to be concentrated in Fraction A of E. hirta aqueous extract.
A comparative study on complete and implant retained denture treatments: a biomechanics perspective

Chen J, Ahmad R, Suenaga H, Li W, Swain M, Li Q

J Biomech, 2015 Feb 5;48(3):512-9.
PMID: 25560272 DOI: 10.1016/j.jbiomech.2014.11.043

Although implant-retained overdenture allows edentulous patients to take higher occlusal forces than the conventional complete dentures, the biomechanical influences have not been explored yet. Clinically, there is limited knowledge and means for predicting localized bone remodelling after denture treatment with and without implant support. By using finite element (FE) analysis, this article provides an in-silico approach to exploring the treatment effects on the oral mucosa and potential resorption of residual ridge under three different denture configurations in a patient-specific manner. Based on cone beam computerized tomography (CBCT) scans, a 3D heterogeneous FE model was created; and the supportive tissue, mucosa, was characterized as a hyperelastic material. A measured occlusal load (63N) was applied onto three virtual models, namely complete denture, two and four implant-retained overdentures. Clinically, the bone resorption was measured after one year in the two implant-retained overdenture treatment. Despite the improved stability and enhanced masticatory function, the implant-retained overdentures demonstrated higher hydrostatic stress in mucosa (43.6kPa and 39.9kPa for two and four implants) at the posterior ends of the mandible due to the cantilever effect, than the complete denture (33.4kPa). Hydrostatic pressure in the mucosa signifies a critical indicator and can be correlated with clinically measured bone resorption, pointing to severer mandibular ridge resorption posteriorly with implant-retained overdentures. This study provides a biomechanical basis for denture treatment planning to improve long-term outcomes with minimal residual ridge resorption.
Interethnic comparisons of important pharmacology genes using SNP databases: potential application to drug regulatory assessments

Chen J, Teo YY, Toh DS, Sung C

Pharmacogenomics, 2010 Aug;11(8):1077-94.
PMID: 20712526 DOI: 10.2217/pgs.10.79

The frequencies of alleles implicated in drug-response variability provide vital information for public health management. Differences in frequencies between genetically diverse groups of individuals can hamper drug assessments, particularly in populations where clinical data are not readily available.
Fulltext Shape Optimization for Additive Manufacturing of Removable Partial Dentures--A New Paradigm for Prosthetic CAD/CAM

Chen J, Ahmad R, Suenaga H, Li W, Sasaki K, Swain M, et al.

PLoS One, 2015;10(7):e0132552.
PMID: 26161878 DOI: 10.1371/journal.pone.0132552

With ever-growing aging population and demand for denture treatments, pressure-induced mucosa lesion and residual ridge resorption remain main sources of clinical complications. Conventional denture design and fabrication are challenged for its labor and experience intensity, urgently necessitating an automatic procedure. This study aims to develop a fully automatic procedure enabling shape optimization and additive manufacturing of removable partial dentures (RPD), to maximize the uniformity of contact pressure distribution on the mucosa, thereby reducing associated clinical complications. A 3D heterogeneous finite element (FE) model was constructed from CT scan, and the critical tissue of mucosa was modeled as a hyperelastic material from in vivo clinical data. A contact shape optimization algorithm was developed based on the bi-directional evolutionary structural optimization (BESO) technique. Both initial and optimized dentures were prototyped by 3D printing technology and evaluated with in vitro tests. Through the optimization, the peak contact pressure was reduced by 70%, and the uniformity was improved by 63%. In vitro tests verified the effectiveness of this procedure, and the hydrostatic pressure induced in the mucosa is well below clinical pressure-pain thresholds (PPT), potentially lessening risk of residual ridge resorption. This proposed computational optimization and additive fabrication procedure provides a novel method for fast denture design and adjustment at low cost, with quantitative guidelines and computer aided design and manufacturing (CAD/CAM) for a specific patient. The integration of digitalized modeling, computational optimization, and free-form fabrication enables more efficient clinical adaptation. The customized optimal denture design is expected to minimize pain/discomfort and potentially reduce long-term residual ridge resorption.
Fulltext Biomechanics of oral mucosa

Chen J, Ahmad R, Li W, Swain M, Li Q

J R Soc Interface, 2015 Aug 06;12(109):20150325.
PMID: 26224566 DOI: 10.1098/rsif.2015.0325

The prevalence of prosthodontic treatment has been well recognized, and the need is continuously increasing with the ageing population. While the oral mucosa plays a critical role in the treatment outcome, the associated biomechanics is not yet fully understood. Using the literature available, this paper provides a critical review on four aspects of mucosal biomechanics, including static, dynamic, volumetric and interactive responses, which are interpreted by its elasticity, viscosity/permeability, apparent Poisson's ratio and friction coefficient, respectively. Both empirical studies and numerical models are analysed and compared to gain anatomical and physiological insights. Furthermore, the clinical applications of such biomechanical knowledge on the mucosa are explored to address some critical concerns, including stimuli for tissue remodelling (interstitial hydrostatic pressure), pressure-pain thresholds, tissue displaceability and residual bone resorption. Through this review, the state of the art in mucosal biomechanics and their clinical implications are discussed for future research interests, including clinical applications, computational modelling, design optimization and prosthetic fabrication.
xDeep-AcPEP: Deep Learning Method for Anticancer Peptide Activity Prediction Based on Convolutional Neural Network and Multitask Learning

Chen J, Cheong HH, Siu SWI

J Chem Inf Model, 2021 Aug 23;61(8):3789-3803.
PMID: 34327990 DOI: 10.1021/acs.jcim.1c00181

Cancer is one of the leading causes of death worldwide. Conventional cancer treatment relies on radiotherapy and chemotherapy, but both methods bring severe side effects to patients, as these therapies not only attack cancer cells but also damage normal cells. Anticancer peptides (ACPs) are a promising alternative as therapeutic agents that are efficient and selective against tumor cells. Here, we propose a deep learning method based on convolutional neural networks to predict biological activity (EC50, LC50, IC50, and LD50) against six tumor cells, including breast, colon, cervix, lung, skin, and prostate. We show that models derived with multitask learning achieve better performance than conventional single-task models. In repeated 5-fold cross validation using the CancerPPD data set, the best models with the applicability domain defined obtain an average mean squared error of 0.1758, Pearson's correlation coefficient of 0.8086, and Kendall's correlation coefficient of 0.6156. As a step toward model interpretability, we infer the contribution of each residue in the sequence to the predicted activity by means of feature importance weights derived from the convolutional layers of the model. The present method, referred to as xDeep-AcPEP, will help to identify effective ACPs in rational peptide design for therapeutic purposes. The data, script files for reproducing the experiments, and the final prediction models can be downloaded from http://github.com/chen709847237/xDeep-AcPEP. The web server to directly access this prediction method is at https://app.cbbio.online/acpep/home.
Fulltext The complete chloroplast genome of Walla Patta, Gyrinops walla (Thymelaeaceae), an agarwood-producing tree species from Sri Lanka

Chen J, Lee SY, Munugoda KD, Mohamed R, Subasinghe SMCUP, Liao W

Mitochondrial DNA B Resour, 2021 May 19;6(6):1699-1701.
PMID: 34104743 DOI: 10.1080/23802359.2021.1926362

Gyrinops walla is an important agarwood-producing tree and threatened species from Sri Lanka. Herein, we assembled and characterized the complete chloroplast (cp) genome of G. walla as a genomic resource for conservation purposes. The 175,130 bp long genome is comprised of 87,376 bp large single-copy (LSC) and 3316 bp small single-copy (SSC) regions, which are separated by two inverted repeat (IR) region, each with a size of 42,291 bp. A total of 140 genes were predicted for the cp genome, which includes 94 protein-coding, 38 tRNA, and eight rRNA genes. Phylogenetic analysis showed that G. walla is fully resolved in a sister position to Aquilaria in the family Thymelaeaceae. The data provided will be useful for study on the molecular phylogenetics and evolution of Thymelaeaceae in the future.
Characterization of Eleusine indica with gene mutation or amplification in EPSPS to glyphosate

Chen J, Jiang C, Huang H, Wei S, Huang Z, Wang H, et al.

Pestic Biochem Physiol, 2017 Nov;143:201-206.
PMID: 29183593 DOI: 10.1016/j.pestbp.2017.09.012

The evolution of weed-resistant species threatens the sustainable use of glyphosate, which is the most important herbicide widely used in agriculture worldwide. Moreover, the high glyphosate resistance (>180-fold based on LD50) of Eleusine indica found in Malaysia, which carries a double mutation in its 5-enolpyruvylshikimate-3-phosphate synthase (EPSPS), made the control of this species more difficult. By contrast, the same species carrying the same double mutation in EPSPS (T102I+P106S) but found in China only shows a resistance level of not more than 14-fold based on GR50. The resistance level of this population is four times higher than that of the population carrying a single mutation (P106L). Although the members of this population survive under a high glyphosate dosage of 10,080gaeha-1, their growth was significantly inhibited by glyphosate under the recommend dose (840gaeha-1), where in the fresh weight was 85.4% of the control. EPSPS expression, relative copy number, and EPSPS activity in this population were similar to those of the susceptible population. In addition, the expression of two glutathione transferase (GST) genes (GST-U8 and GST-23) and the enzyme activity of the GST in this population did not significantly differ from those of the susceptible population. This finding is important in elucidating the resistance of the naturally evolved glyphosate-resistant (GR) weed species carrying a double mutation in EPSPS to glyphosate.
Fulltext Chemical toxicity prediction based on semi-supervised learning and graph convolutional neural network

Chen J, Si YW, Un CW, Siu SWI

J Cheminform, 2021 Nov 27;13(1):93.
PMID: 34838140 DOI: 10.1186/s13321-021-00570-8

As safety is one of the most important properties of drugs, chemical toxicology prediction has received increasing attentions in the drug discovery research. Traditionally, researchers rely on in vitro and in vivo experiments to test the toxicity of chemical compounds. However, not only are these experiments time consuming and costly, but experiments that involve animal testing are increasingly subject to ethical concerns. While traditional machine learning (ML) methods have been used in the field with some success, the limited availability of annotated toxicity data is the major hurdle for further improving model performance. Inspired by the success of semi-supervised learning (SSL) algorithms, we propose a Graph Convolution Neural Network (GCN) to predict chemical toxicity and trained the network by the Mean Teacher (MT) SSL algorithm. Using the Tox21 data, our optimal SSL-GCN models for predicting the twelve toxicological endpoints achieve an average ROC-AUC score of 0.757 in the test set, which is a 6% improvement over GCN models trained by supervised learning and conventional ML methods. Our SSL-GCN models also exhibit superior performance when compared to models constructed using the built-in DeepChem ML methods. This study demonstrates that SSL can increase the prediction power of models by learning from unannotated data. The optimal unannotated to annotated data ratio ranges between 1:1 and 4:1. This study demonstrates the success of SSL in chemical toxicity prediction; the same technique is expected to be beneficial to other chemical property prediction tasks by utilizing existing large chemical databases. Our optimal model SSL-GCN is hosted on an online server accessible through: https://app.cbbio.online/ssl-gcn/home .

Filters

Please provide feedback to Administrator (afdal@afpm.org.my)

External Links