Publications by authors named "Arnaud Droit"

109 Publications

Phenotypic Trade-Offs: Deciphering the Impact of Neurodiversity on Drug Development in Fragile X Syndrome.

Front Psychiatry 2021 18;12:730987. Epub 2021 Oct 18.

Department of Pediatrics, University of Alberta, Edmonton, AB, Canada.

Fragile X syndrome (FXS) is the most common single-gene cause of intellectual disability and autism spectrum disorder. Individuals with FXS present with a wide range of severity in multiple phenotypes including cognitive delay, behavioral challenges, sleep issues, epilepsy, and anxiety. These symptoms are also shared by many individuals with other neurodevelopmental disorders (NDDs). Since the discovery of the FXS gene, FMR1, FXS has been the focus of intense preclinical investigation and is placed at the forefront of clinical trials in the field of NDDs. So far, most studies have aimed to translate the rescue of specific phenotypes in animal models, for example, learning, or improving general cognitive or behavioral functioning in individuals with FXS. Trial design, selection of outcome measures, and interpretation of results of recent trials have shown limitations in this type of approach. We propose a new paradigm in which all phenotypes involved in individuals with FXS would be considered and, more importantly, the possible interactions between these phenotypes. This approach would be implemented both at the baseline, meaning when entering a trial or when studying a patient population, and also after the intervention when the study subjects have been exposed to the investigational product. This approach would allow us to further understand potential trade-offs underlying the varying effects of the treatment on different individuals in clinical trials, and to connect the results to individual genetic differences. To better understand the interplay between different phenotypes, we emphasize the need for preclinical studies to investigate various interrelated biological and behavioral outcomes when assessing a specific treatment. In this paper, we present how such a conceptual shift in preclinical design could shed new light on clinical trial results. Future clinical studies should take into account the rich neurodiversity of individuals with FXS specifically and NDDs in general, and incorporate the idea of trade-offs in their designs.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3389/fpsyt.2021.730987DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8558248PMC
October 2021

Proteomic Analysis of Maternal Urine for the Early Detection of Preeclampsia and Fetal Growth Restriction.

J Clin Med 2021 Oct 13;10(20). Epub 2021 Oct 13.

CHU de Québec-Université Laval Research Center, Université Laval, Quebec City, QC G1V 4G2, Canada.

Background: To explore the use of maternal urine proteome for the identification of preeclampsia biomarkers.

Methods: Maternal urine samples from women with and without preeclampsia were used for protein discovery followed by a validation study. The targeted proteins of interest were then measured in urine samples collected at 20-24 and 30-34 weeks among nine women who developed preeclampsia, one woman with fetal growth restriction, and 20 women with uncomplicated pregnancies from a longitudinal study. Protein identification and quantification was obtained using liquid chromatography-tandem mass spectrometry (LC-MS/MS).

Results: Among the 1108 urine proteins quantified in the discovery study, 21 were upregulated in preeclampsia and selected for validation. Nineteen (90%) proteins were confirmed as upregulated in preeclampsia cases. Among them, two proteins, ceruloplasmin and serpin A7, were upregulated at 20-24 weeks and 30-34 weeks of gestation ( < 0.05) in cases of preeclampsia, and could have served to identify 60% of women who subsequently developed preeclampsia and/or fetal growth restriction at 20-24 weeks of gestation, and 78% at 30-34 weeks, for a false-positive rate of 10%.

Conclusions: Proteomic profiling of maternal urine can differentiate women with and without preeclampsia. Several proteins including ceruloplasmin and serpin A7 are upregulated in maternal urine before the diagnosis of preeclampsia and potentially fetal growth restriction.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3390/jcm10204679DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8537852PMC
October 2021

Vitamin C Differentially Impacts the Serum Proteome Profile in Female and Male Mice.

J Proteome Res 2021 Nov 13;20(11):5036-5053. Epub 2021 Oct 13.

Centre de recherche du CHU de Québec, Faculty of Medicine, Université Laval, Québec City, Québec G1 V 4G2, Canada.

A suboptimal blood vitamin C (ascorbate) level increases the risk of several chronic diseases. However, the detection of hypovitaminosis C is not a simple task, as ascorbate is unstable in blood samples. In this study, we examined the serum proteome of mice lacking the gulonolactone oxidase (Gulo) required for the ascorbate biosynthesis. mice were supplemented with different concentrations of ascorbate in drinking water, and serum was collected to identify proteins correlating with serum ascorbate levels using an unbiased label-free liquid chromatography-tandem mass spectrometry global quantitative proteomic approach. Parallel reaction monitoring was performed to validate the correlations. We uncovered that the serum proteome profiles differ significantly between male and female mice. Also, unlike males, a four-week ascorbate treatment did not entirely re-establish the serum proteome profile of ascorbate-deficient females to the optimal profile exhibited by females that never experienced an ascorbate deficiency. Finally, the serum proteins involved in retinoid metabolism, cholesterol, and lipid transport were similarly affected by ascorbate levels in males and females. In contrast, the proteins regulating serum peptidases and the protein of the acute phase response were different between males and females. These proteins are potential biomarkers correlating with blood ascorbate levels and require further study in standard clinical settings. The complete proteomics data set generated in this study has been deposited to the public repository ProteomeXchange with the data set identifier: PXD027019.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1021/acs.jproteome.1c00542DOI Listing
November 2021

Identification of Abundant and Functional dodecaRNAs (doRNAs) Derived from Ribosomal RNA.

Int J Mol Sci 2021 Sep 9;22(18). Epub 2021 Sep 9.

CHU de Québec Research Center/CHUL Pavilion-Université Laval, 2705 boulevard Laurier, Quebec City, QC G1V 4G2, Canada.

Using a modified RNA-sequencing (RNA-seq) approach, we discovered a new family of unusually short RNAs mapping to ribosomal RNA 5.8S, which we named dodecaRNAs (doRNAs), according to the number of core nucleotides (12 nt) their members contain. Using a new quantitative detection method that we developed, we confirmed our RNA-seq data and determined that the minimal core doRNA sequence and its 13-nt variant C-doRNA (doRNA with a 5' Cytosine) are the two most abundant doRNAs, which, together, may outnumber microRNAs. The C-doRNA/doRNA ratio is stable within species but differed between species. doRNA and C-doRNA are mainly cytoplasmic and interact with heterogeneous nuclear ribonucleoproteins (hnRNP) A0, A1 and A2B1, but not Argonaute 2. Reporter gene activity assays suggest that C-doRNA may function as a regulator of Annexin II receptor (AXIIR) expression. doRNAs are differentially expressed in prostate cancer cells/tissues and may control cell migration. These findings suggest that unusually short RNAs may be more abundant and important than previously thought.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3390/ijms22189757DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8467515PMC
September 2021

timeOmics: an R package for longitudinal multi-omics data integration.

Bioinformatics 2021 Sep 23. Epub 2021 Sep 23.

Molecular Medicine Department, CHU de Québec Research Center, Université Laval, Québec, QC, Canada.

Motivation: Multi-omics data integration enables the global analysis of biological systems and discovery of new biological insights. Multi-omics experimental designs have been further extended with a longitudinal dimension to study dynamic relationships between molecules. However, methods that integrate longitudinal multi-omics data are still in their infancy.

Results: We introduce the R package timeOmics, a generic analytical framework for the integration of longitudinal multi-omics data. The framework includes pre-processing, modelling and clustering to identify molecular features strongly associated with time. We illustrate this framework in a case study to detect seasonal patterns of mRNA, metabolites, gut taxa, and clinical variables in patients with diabetes mellitus from the integrative Human Microbiome Project.

Availability: timeOmics is available on Bioconductor and github.com/abodein/timeOmics.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1093/bioinformatics/btab664DOI Listing
September 2021

The gut microbiome in konzo.

Nat Commun 2021 09 10;12(1):5371. Epub 2021 Sep 10.

Center for Genetic Medicine Research, Children's Research Institute, Children's National Hospital, Washington, DC, USA.

Konzo, a distinct upper motor neuron disease associated with a cyanogenic diet and chronic malnutrition, predominately affects children and women of childbearing age in sub-Saharan Africa. While the exact biological mechanisms that cause this disease have largely remained elusive, host-genetics and environmental components such as the gut microbiome have been implicated. Using a large study population of 180 individuals from the Democratic Republic of the Congo, where konzo is most frequent, we investigate how the structure of the gut microbiome varied across geographical contexts, as well as provide the first insight into the gut flora of children affected with this debilitating disease using shotgun metagenomic sequencing. Our findings indicate that the gut microbiome structure is highly variable depending on region of sampling, but most interestingly, we identify unique enrichments of bacterial species and functional pathways that potentially modulate the susceptibility of konzo in prone regions of the Congo.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1038/s41467-021-25694-1DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8433213PMC
September 2021

Extensive and Accurate Benchmarking of DIA Acquisition Methods and Software Tools Using a Complex Proteomic Standard.

J Proteome Res 2021 Oct 2;20(10):4801-4814. Epub 2021 Sep 2.

Proteomics Platform, CHU de Québec - Université Laval Research Centre, Québec City, Québec G1V 4G2, Canada.

Over the past decade, the data-independent acquisition mode has gained popularity for broad coverage of complex proteomes by LC-MS/MS and quantification of low-abundance proteins. However, there is no consensus in the literature on the best data acquisition parameters and processing tools to use for this specific application. Here, we present the most comprehensive comparison of DIA workflows on Orbitrap instruments published so far in the field of proteomics. Using a standard human 48 proteins mixture (UPS1-Sigma) at 8 different concentrations in an proteome background, we tested 36 workflows including 4 different DIA window acquisition schemes and 6 different software tools (DIA-NN, DIA-Umpire, OpenSWATH, ScaffoldDIA, Skyline, and Spectronaut) with or without the use of a DDA spectral library. On the basis of the number of proteins identified, quantification linearity and reproducibility, as well as sensitivity and specificity in 28 pairwise comparisons of different UPS1 concentrations, we summarize the major considerations and propose guidelines for choosing the DIA workflow best suited for LC-MS/MS proteomic analyses. Our 96 DIA raw files and software outputs have been deposited on ProteomeXchange for testing or developing new DIA processing tools.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1021/acs.jproteome.1c00490DOI Listing
October 2021

Lipoprotein Proteomics and Aortic Valve Transcriptomics Identify Biological Pathways Linking Lipoprotein(a) Levels to Aortic Stenosis.

Metabolites 2021 Jul 16;11(7). Epub 2021 Jul 16.

Centre de Recherche de l'Institut Universitaire de Cardiologie et de Pneumologie de Québec, Québec, QC G1V 4G5, Canada.

Lipoprotein(a) (Lp(a)) is one of the most important risk factors for the development of calcific aortic valve stenosis (CAVS). However, the mechanisms through which Lp(a) causes CAVS are currently unknown. Our objectives were to characterize the Lp(a) proteome and to identify proteins that may be differentially associated with Lp(a) in patients with versus without CAVS. Our second objective was to identify genes that may be differentially regulated by exposure to high versus low Lp(a) levels in explanted aortic valves from patients with CAVS. We isolated Lp(a) from the blood of 21 patients with CAVS and 22 volunteers and performed untargeted label-free analysis of the Lp(a) proteome. We also investigated the transcriptomic signature of calcified aortic valves from patients who underwent aortic valve replacement with high versus low Lp(a) levels ( = 118). Proteins involved in the protein activation cascade, platelet degranulation, leukocyte migration, and response to wounding may be associated with Lp(a) depending on CAVS status. The transcriptomic analysis identified genes involved in cardiac aging, chondrocyte development, and inflammation as potentially influenced by Lp(a). Our multi-omic analyses identified biological pathways through which Lp(a) may cause CAVS, as well as key molecular events that could be triggered by Lp(a) in CAVS development.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3390/metabo11070459DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8307014PMC
July 2021

Integration strategies of multi-omics data for machine learning analysis.

Comput Struct Biotechnol J 2021 22;19:3735-3746. Epub 2021 Jun 22.

Molecular Medicine Department, CHU de Québec Research Center, Université Laval, Québec, QC, Canada.

Increased availability of high-throughput technologies has generated an ever-growing number of omics data that seek to portray many different but complementary biological layers including genomics, epigenomics, transcriptomics, proteomics, and metabolomics. New insight from these data have been obtained by machine learning algorithms that have produced diagnostic and classification biomarkers. Most biomarkers obtained to date however only include one omic measurement at a time and thus do not take full advantage of recent multi-omics experiments that now capture the entire complexity of biological systems. Multi-omics data integration strategies are needed to combine the complementary knowledge brought by each omics layer. We have summarized the most recent data integration methods/ frameworks into five different integration strategies: early, mixed, intermediate, late and hierarchical. In this mini-review, we focus on challenges and existing multi-omics integration strategies by paying special attention to machine learning applications.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.csbj.2021.06.030DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8258788PMC
June 2021

A Homozygous Deep Intronic Mutation Alters the Splicing of Nebulin Gene in a Patient With Nemaline Myopathy.

Front Neurol 2021 15;12:660113. Epub 2021 Jun 15.

Centre de recherche CHU de Québec- Laval University, Quebec City, QC, Canada.

Nemaline myopathy is a rare disorder affecting the muscle sarcomere. Mutations in nebulin gene () are known to be responsible for about 50% of nemaline myopathy cases. Nebulin is a giant protein which is formed integrally with the sarcomeric thin filament. This complex gene is under extensive alternative splicing giving rise to multiple isoforms. In this study, we report a 6-year-old boy presenting with general muscular weaknesses. Identification of rod-shaped structures in the patient' biopsy raised doubt about the presence of a nemaline myopathy. Next-generation sequencing was used to identify a causative mutation for the patient syndrome. A homozygous deep intronic substitution was found in the intron 144 of the . The variant was predicted by tools to create a new donor splice site. Molecular analysis has shown that the mutation could alter splicing events of the nebulin gene leading to a significant decrease of isoforms level. This change in the expression level of nebulin could give rise to functional consequences in the sarcomere. These results are consistent with the phenotypes observed in the patient. Such a discovery of variants in this gene will allow a better understanding of the involvement of nebulin in neuromuscular diseases and help find new treatments for the nemaline myopathy.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3389/fneur.2021.660113DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8239344PMC
June 2021

Personalized Risk Assessment for Prevention and Early Detection of Breast Cancer: Integration and Implementation (PERSPECTIVE I&I).

J Pers Med 2021 Jun 4;11(6). Epub 2021 Jun 4.

CHU de Québec-Université Laval Research Center, Québec City, QC G1V 4G2, Canada.

Early detection of breast cancer through screening reduces breast cancer mortality. The benefits of screening must also be considered within the context of potential harms (e.g., false positives, overdiagnosis). Furthermore, while breast cancer risk is highly variable within the population, most screening programs use age to determine eligibility. A risk-based approach is expected to improve the benefit-harm ratio of breast cancer screening programs. The PERSPECTIVE I&I (Personalized Risk Assessment for Prevention and Early Detection of Breast Cancer: Integration and Implementation) project seeks to improve personalized risk assessment to allow for a cost-effective, population-based approach to risk-based screening and determine best practices for implementation in Canada. This commentary describes the four inter-related activities that comprise the PERSPECTIVE I&I project. 1: Identification and validation of novel moderate to high-risk susceptibility genes. 2: Improvement, validation, and adaptation of a risk prediction web-tool for the Canadian context. 3: Development and piloting of a socio-ethical framework to support implementation of risk-based breast cancer screening. 4: Economic analysis to optimize the implementation of risk-based screening. Risk-based screening and prevention is expected to benefit all women, empowering them to work with their healthcare provider to make informed decisions about screening and prevention.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3390/jpm11060511DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8226444PMC
June 2021

GWENA: gene co-expression networks analysis and extended modules characterization in a single Bioconductor package.

BMC Bioinformatics 2021 May 25;22(1):267. Epub 2021 May 25.

Département de médecine moléculaire, Faculté de médecine, Université Laval, 2325 rue de l'Université, Québec, G1V 0A6, Canada.

Background: Network-based analysis of gene expression through co-expression networks can be used to investigate modular relationships occurring between genes performing different biological functions. An extended description of each of the network modules is therefore a critical step to understand the underlying processes contributing to a disease or a phenotype. Biological integration, topology study and conditions comparison (e.g. wild vs mutant) are the main methods to do so, but to date no tool combines them all into a single pipeline.

Results: Here we present GWENA, a new R package that integrates gene co-expression network construction and whole characterization of the detected modules through gene set enrichment, phenotypic association, hub genes detection, topological metric computation, and differential co-expression. To demonstrate its performance, we applied GWENA on two skeletal muscle datasets from young and old patients of GTEx study. Remarkably, we prioritized a gene whose involvement was unknown in the muscle development and growth. Moreover, new insights on the variations in patterns of co-expression were identified. The known phenomena of connectivity loss associated with aging was found coupled to a global reorganization of the relationships leading to expression of known aging related functions.

Conclusion: GWENA is an R package available through Bioconductor ( https://bioconductor.org/packages/release/bioc/html/GWENA.html ) that has been developed to perform extended analysis of gene co-expression networks. Thanks to biological and topological information as well as differential co-expression, the package helps to dissect the role of genes relationships in diseases conditions or targeted phenotypes. GWENA goes beyond existing packages that perform co-expression analysis by including new tools to fully characterize modules, such as differential co-expression, additional enrichment databases, and network visualization.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1186/s12859-021-04179-4DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8152313PMC
May 2021

A Comparative Analysis of the Lipoprotein(a) and Low-Density Lipoprotein Proteomic Profiles Combining Mass Spectrometry and Mendelian Randomization.

CJC Open 2021 Apr 3;3(4):450-459. Epub 2020 Dec 3.

Centre de recherche de l'Institut universitaire de cardiologie et de pneumologie de Québec, Quebec, Canada.

Background: Lipoprotein(a) (Lp[a]), which consists of a low-density lipoprotein (LDL) bound to apolipoprotein(a), is one of the strongest genetic risk factors for atherosclerotic cardiovascular diseases. Few studies have performed hypothesis-free direct comparisons of the Lp(a) and the LDL proteomes. Our objectives were to compare the Lp(a) and the LDL proteomic profiles and to evaluate the effect of lifelong exposure to elevated Lp(a) or LDL cholesterol levels on the plasma proteomic profile.

Methods: We performed a label-free analysis of the Lp(a) and LDL proteomic profiles of healthy volunteers in a discovery (n = 6) and a replication (n = 9) phase. We performed inverse variance weighted Mendelian randomization to document the effect of lifelong exposure to elevated Lp(a) or LDL cholesterol levels on the plasma proteomic profile of participants of the INTERVAL study.

Results: We identified 15 proteins that were more abundant on Lp(a) compared with LDL (, , , , , , , , , , , , , , and ). We found no proteins that were more abundant on LDL compared with Lp(a). After correction for multiple testing, lifelong exposure to elevated LDL cholesterol levels was associated with the variation of 18 plasma proteins whereas Lp(a) did not appear to influence the plasma proteome.

Conclusions: Results of this study highlight marked differences in the proteome of Lp(a) and LDL as well as in the effect of lifelong exposure to elevated LDL cholesterol or Lp(a) on the plasma proteomic profile.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.cjco.2020.11.019DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8129481PMC
April 2021

HCK and ABAA: A Newly Designed Pipeline to Improve Fungi Metabarcoding Analysis.

Front Microbiol 2021 5;12:640693. Epub 2021 May 5.

Department of Molecular Medicine, Laval University, Quebec, QC, Canada.

Introduction: The fungi ITS sequence length dissimilarity, non-specific amplicons, including chimaera formed during Polymerase Chain Reaction (PCR), added to sequencing errors, create bias during similarity clustering and abundance estimation in the downstream analysis. To overcome these challenges, we present a novel approach, Hierarchical Clustering with Kraken (HCK), to classify ITS1 amplicons and Abundance-Base Alternative Approach (ABAA) pipeline to detect and filter non-specific amplicons in fungi metabarcoding sequencing datasets.

Materials And Methods: We compared the performances of both pipelines against QIIME, KRAKEN, and DADA2 using publicly available fungi ITS mock community datasets and using BLASTn as a reference. We calculated the Precision, Recall, F-score using the True-Positive, False-positive, and False-negative estimation. Alpha diversity (Chao1 and Shannon metrics) was also used to evaluate the diversity estimation of our method.

Results: The analysis shows that ABAA reduced the number of false-positive with all metabarcoding methods tested, and HCK increases precision and recall. HCK, coupled with ABAA, improves the F-score and bring alpha diversity metric value close to that of the BLASTn alpha diversity values when compared to QIIME, KRAKEN, and DADA2.

Conclusion: The developed HCK-ABAA approach allows better identification of the fungi community structures while avoiding use of a reference database for non-specific amplicons filtration. It results in a more robust and stable methodology over time. The software can be downloaded on the following link: https://bitbucket.org/GottySG36/hck/src/master/.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3389/fmicb.2021.640693DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8134036PMC
May 2021

Differential Inhibition of HIV Replication by the 12 Interferon Alpha Subtypes.

J Virol 2021 07 12;95(15):e0231120. Epub 2021 Jul 12.

INSERM UMR-1124, Paris, France.

Type I interferons (IFNs) are a family of cytokines that represent a first line of defense against virus infections. The 12 different IFN-α subtypes share a receptor on target cells and trigger similar signaling cascades. Several studies have collectively shown that this apparent redundancy conceals qualitatively different responses induced by individual subtypes, which display different efficacies of inhibition of HIV replication. Some studies, however, provided evidence that the disparities are quantitative rather than qualitative. Since RNA expression analyses show a large but incomplete overlap of the genes induced, they may support both models. To explore if the IFN-α subtypes induce functionally relevant different anti-HIV activities, we have compared the efficacies of inhibition of all 12 subtypes on HIV spread and on specific steps of the viral replication cycle, including viral entry, reverse transcription, protein synthesis, and virus release. Finding different hierarchies of inhibition would validate the induction of qualitatively different responses. We found that while most subtypes similarly inhibit virus entry, they display distinctive potencies on other early steps of HIV replication. In addition, only some subtypes were able to target effectively the late steps. The extent of induction of known anti-HIV factors helps to explain some, but not all differences observed, confirming the participation of additional IFN-induced anti-HIV effectors. Our findings support the notion that different IFN-α subtypes can induce the expression of qualitatively different antiviral activities. The initial response against viruses relies in large part on type I interferons, which include 12 subtypes of IFN-α. These cytokines bind to a common receptor on the cell surface and trigger the expression of incompletely overlapping sets of genes. Whether the anti-HIV responses induced by IFN-α subtypes differ in the extent of expression or in the nature of the genes involved remains debated. Also, RNA expression profiles led to opposite conclusions, depending on the importance attributed to the induction of common or distinctive genes. To explore if relevant anti-HIV activities can be differently induced by the IFN-α subtypes, we compared their relative efficacies on specific steps of the replication cycle. We show that the hierarchy of IFN potencies depends on the step analyzed, supporting qualitatively different responses. This work will also prompt the search for novel IFN-induced anti-HIV factors acting on specific steps of the replication cycle.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1128/JVI.02311-20DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8274621PMC
July 2021

A founder mutation in the gene in families from Saguenay-Lac-St-Jean region affected by a pyridoxine-dependent epilepsy.

JIMD Rep 2021 May 23;59(1):32-41. Epub 2021 Feb 23.

Centre de recherche CHU de Québec- Université Laval, Laval University Québec Québec Canada.

Pyridoxine-dependent epilepsy (PDE) is a relatively rare subgroup of epileptic disorders. They generally present in infancy as an early onset epileptic encephalopathy or seizures, refractory to standard treatments, with rapid and variable responses to vitamin B6 treatment. Whole exome sequencing of three unrelated families identified homozygous pathogenic mutation c.370_373del, p.Asp124fs in gene in five persons. Haplotype analysis showed a single shared profile for the affected persons and their parents, leading to a hypothesis about founder effect of the mutation in Saguenay-Lac-St-Jean region of French Canadians. All affected probands also shared one single mitochondrial haplotype T2b3 and two rare variations in the mitochondrial genome m.801A>G and m.5166A>G suggesting that a single individual female introduced mutation c.370_373del, p.Asp124fs in Quebec. The mutation p.Asp124fs causes a severe disease phenotype with delayed myelination and cortical/subcortical brain atrophy. The most noteworthy radiological finding in this Quebec founder mutation is the presence of the temporal cysts that can be used as a marker of the disease. Also, both patients, who are alive, had a history of prenatal supplements taken by their mothers as antiemetic medication with high doses of pyridoxine. In the context of suspected PDE in patients with neonatal refractory seizures, treatment with pyridoxine and/or Pyridoxal-5-phophate has to be started immediately and continued until the results of genetic analysis received. Even with early appropriate treatment, neurological outcome of our patient is still poor.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1002/jmd2.12196DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8100403PMC
May 2021

KibioR & Kibio: a new architecture for next-generation data querying and sharing in big biology.

Bioinformatics 2021 Mar 4. Epub 2021 Mar 4.

Centre de Recherche du CHU de Québec - Université Laval, Québec, Québec, Canada.

Motivation: The growing production of massive heterogeneous biological data offers opportunities for new discoveries. However, performing multi-omics data analysis is challenging, and researchers are forced to handle the ever-increasing complexity of both data management and evolution of our biological understanding. Substantial efforts have been made to unify biological datasets into integrated systems. Unfortunately, they are not easily scalable, deployable and searchable, locally or globally.

Results: This publication presents two tools with a simple structure that can help any data provider, organization or researcher, requiring a reliable data search and analysis base. The first tool is Kibio, a scalable and adaptable data storage based on Elasticsearch search engine. The second tool is KibioR, a R package to pull, push and search Kibio datasets or any accessible Elasticsearch-based databases. These tools apply a uniform data exchange model and minimize the burden of data management by organizing data into a decentralized, versatile, searchable and shareable structure. Several case studies are presented using multiple databases, from drug characterization to miRNAs and pathways identification, emphasizing the ease of use and versatility of the Kibio/KibioR framework.

Availability: Both KibioR and Elasticsearch are open source. KibioR package source is available at https://github.com/regisoc/kibior and the library on CRAN at https://cran.r-project.org/package=kibior.

Supplementary Information: Supplementary data are available at Bioinformatics online.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1093/bioinformatics/btab157DOI Listing
March 2021

Machine Learning-Based Individualized Survival Prediction Model for Total Knee Replacement in Osteoarthritis: Data From the Osteoarthritis Initiative.

Arthritis Care Res (Hoboken) 2021 10 26;73(10):1518-1527. Epub 2021 Aug 26.

Laval University Hospital Research Centre, Quebec, Canada.

Objective: By using machine learning, our study aimed to build a model to predict risk and time to total knee replacement (TKR) of an osteoarthritic knee.

Methods: Features were from the Osteoarthritis Initiative (OAI) cohort at baseline. Using the lasso method for variable selection in the Cox regression model, we identified the 10 most important characteristics among 1,107 features. The prognostic power of the selected features was assessed by the Kaplan-Meier method and applied to 7 machine learning methods: Cox, DeepSurv, random forests algorithm, linear/kernel support vector machine (SVM), and linear/neural multi-task logistic regression models. As some of the 10 first-found features included similar radiographic measurements, we further looked at using the least number of features without compromising the accuracy of the model. Prediction performance was assessed by the concordance index, Brier score, and time-dependent area under the curve (AUC).

Results: Ten features were identified and included radiographs, bone marrow lesions of the medial condyle on magnetic resonance imaging, hyaluronic acid injection, performance measure, medical history, and knee-related symptoms. The methodologies Cox, DeepSurv, and linear SVM demonstrated the highest accuracy (concordance index scores of 0.85, Brier score of 0.02, and an AUC of 0.87). DeepSurv was chosen to build the prediction model to estimate the time to TKR for a given knee. Moreover, we were able to decrease the features to only 3 and maintain the high accuracy (concordance index of 0.85, Brier score of 0.02, and AUC of 0.86), which included bone marrow lesions, Kellgren/Lawrence grade, and knee-related symptoms, to predict risk and time of a TKR event.

Conclusion: For the first time, we developed a model using the OAI cohort to predict with high accuracy if a given osteoarthritic knee would require TKR, when a TKR would be required, and who would likely progress fast toward this event.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1002/acr.24601DOI Listing
October 2021

Male sex chromosomal complement exacerbates the pathogenicity of Th17 cells in a chronic model of central nervous system autoimmunity.

Cell Rep 2021 03;34(10):108833

axe Neurosciences, Centre de Recherche du CHU de Québec-Université Laval, Pavillon CHUL, 2705 boulevard Laurier, Quebec City, QC G1V 4G2, Canada; Faculty of Medicine, Université Laval, 1050 ave de la Médecine, Quebec City, QC, Canada. Electronic address:

Sex differences in multiple sclerosis (MS) incidence and severity have long been recognized. However, the underlying cellular and molecular mechanisms for why male sex is associated with more aggressive disease remain poorly defined. Using a T cell adoptive transfer model of chronic experimental autoimmune encephalomyelitis (EAE), we find that male Th17 cells induce disease of increased severity relative to female Th17 cells, irrespective of whether transferred to male or female recipients. Throughout the disease course, a greater frequency of male Th17 cells produce IFNγ, a hallmark of pathogenic Th17 responses. Intriguingly, XY chromosomal complement increases the pathogenicity of male Th17 cells. An X-linked immune regulator, Jarid1c, is downregulated in pathogenic male murine Th17 cells, and functional experiments reveal that it represses the severity of Th17-mediated EAE. Furthermore, Jarid1c expression is downregulated in CD4 T cells from MS-affected individuals. Our data indicate that male sex chromosomal complement critically regulates Th17 cell pathogenicity.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.celrep.2021.108833DOI Listing
March 2021

KLF5 and NFYA factors as novel regulators of prostate cancer cell metabolism.

Endocr Relat Cancer 2021 Apr;28(4):257-271

Département de médecine moléculaire, Faculté de médecine, Axe Endocrinologie - Néphrologie du Centre de recherche Centre Hospitalier Universitaire (CHU) de Québec - Université Laval, et Centre de recherche sur le cancer - Université Laval, Québec, Canada.

Prostate cancer (PCa) cells rely on the androgen receptor (AR) signaling axis to reprogram metabolism to sustain aberrant proliferation. Whether additional transcription factors participate to this reprogramming remains mostly unknown. To identify such factors, DNA motif analyses were performed in the promoter and regulatory regions of genes sensitive to androgens in PCa cells. These analyses identified two transcription factors, KLF5 and NFYA, as possibly associated with PCa cell metabolism. In clinical datasets, KLF5 and NFYA expression levels were associated with disease aggressiveness, being significantly decreased and increased, respectively, during PCa progression. Their expression was next investigated by qPCR and Western blot in human PCa cell models, revealing a positive regulation of KLF5 by androgens and a correlation between NFYA and AR protein expression status. siRNA-mediated knockdown of KLF5 increased human PCa cell proliferation rate in AR-positive cell models, suggesting a tumor suppressor function. Live-cell metabolic assays showed that knockdown of KLF5 promoted mitochondrial respiration, a key metabolic pathway associated with PCa progression. The opposite was observed for knockdown of NFYA regarding proliferation and respiration. RNA-seq analyses following the knockdown of either KLF5 and NFYA confirmed that both factors regulated distinct metabolic gene signatures, as well as other gene signatures, explaining their differential impact on PCa cell proliferation and metabolism. Overall, our findings identify KLF5 and NFYA as novel regulators of PCa cell metabolism.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1530/ERC-20-0504DOI Listing
April 2021

Mapping the Human Herpesvirus 6B transcriptome.

J Virol 2021 Feb 24. Epub 2021 Feb 24.

Division of Infectious Disease and Immunity, CHU de Québec Research Center, Quebec City, Québec, Canada

The "omics" revolution of recent years has simplified the study of RNA transcripts produced during viral infection and under specific defined conditions. In the quest to find new and differentially expressed transcripts during the course of human Herpesvirus 6B (HHV-6B) infection, we made use of large-scale RNA sequencing to analyze the HHV-6B transcriptome during productive infection of human Molt-3 T-cells. Analyses were performed at different time points following infection and specific inhibitors were used to classify the kinetic class of each open reading frame (ORF) reported in the annotated genome of HHV-6B Z29 strain. The initial search focussed on HHV-6B-specific reads matching new HHV-6B transcripts. Differential expression of new HHV-6B transcripts were observed in all samples analyzed. The presence of many of these new HHV-6B transcripts were confirmed by RT-PCR and Sanger sequencing. Many of these transcripts represented new splice variants of previously reported ORFs, including some transcripts that have yet to be defined. Overall, our work demonstrates the diversity and the complexity of the HHV-6B transcriptome.RNA sequencing (RNA-seq) is an important tool for studying RNA transcripts, particularly during active viral infection. We made use of RNA-seq to study human Herpesvirus 6B (HHV-6B) infection. Using six different time points, we were able to identify the presence of differentially spliced genes at 6, 9, 12, 24, 48 and 72 hours post-infection. Determination of the RNA profiles in the presence of cycloheximide (CHX) or phosphonoacetic acid (PAA) also permitted identification of the kinetic class of each ORF described in the annotated GenBank file. We also identified new spliced transcripts for certain genes and evaluated their relative expression over time. These data and next-generation sequencing (NGS) of the viral DNA have led us to propose a new version of the HHV-6B Z29 GenBank annotated file, without changing ORF names in order to facilitate trace back and correlate our work with previous studies on HHV-6B.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1128/JVI.01335-20DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8139660PMC
February 2021

Modulating HSF1 levels impacts expression of the estrogen receptor α and antiestrogen response.

Life Sci Alliance 2021 05 16;4(5). Epub 2021 Feb 16.

Centre de Recherche du CHU de Québec - Université Laval, Axe Oncologie, Québec, Canada

Master transcription factors control the transcriptional program and are essential to maintain cellular functions. Among them, steroid nuclear receptors, such as the estrogen receptor α (ERα), are central to the etiology of hormone-dependent cancers which are accordingly treated with corresponding endocrine therapies. However, resistance invariably arises. Here, we show that high levels of the stress response master regulator, the heat shock factor 1 (HSF1), are associated with antiestrogen resistance in breast cancer cells. Indeed, overexpression of HSF1 leads to ERα degradation, decreased expression of ERα-activated genes, and antiestrogen resistance. Furthermore, we demonstrate that reducing HSF1 levels reinstates expression of the ERα and restores response to antiestrogens. Last, our results establish a proof of concept that inhibition of HSF1, in combination with antiestrogens, is a valid strategy to tackle resistant breast cancers. Taken together, we are proposing a mechanism where high HSF1 levels interfere with the ERα-dependent transcriptional program leading to endocrine resistance in breast cancer.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.26508/lsa.202000811DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7893817PMC
May 2021

ETumorMetastasis: A Network-based Algorithm Predicts Clinical Outcomes Using Whole-exome Sequencing Data of Cancer Patients.

Genomics Proteomics Bioinformatics 2021 Feb 10. Epub 2021 Feb 10.

Human Health Therapeutics, National Research Council Canada, Montreal H4P 2R2, Canada; Department of Biochemistry & Molecular Biology, Medical Genetics, and Oncology, University of Calgary, Calgary T2N 4N1, Canada; Alberta Children's Hospital Research Institute and Arnie Charbonneau Cancer Research Institute, University of Calgary, Calgary T2N 4N1, Canada; Department of Medicine, McGill University, Montreal H3G 2M1, Canada. Electronic address:

Continual reduction in sequencing cost is expanding the accessibility of genome sequencing data for routine clinical applications. However, the lack of methods to construct machine learning-based predictive models using these datasets has become a crucial bottleneck for the application of sequencing technology in clinics. Here we developed a new algorithm, eTumorMetastasis, which transforms tumor functional mutations into network-based profiles and identifies network operational gene signatures (NOG signatures). NOG signatures model the tipping point at which a tumor cell shifts from a state that doesn't favor recurrence to one that does. We showed that NOG signatures derived from genomic mutations of tumor founding clones (i.e., the 'most recent common ancestor' of the cells within a tumor) significantly distinguished recurred and non-recurred breast tumors as well as outperformed the most popular genomic test (i.e., Oncotype DX breast cancer recurrence score). These results imply that mutations of the tumor founding clones are associated with tumor recurrence and can be used to predict clinical outcomes. As such, predictive tools could be used in clinics to guide treatment routes. Finally, the concepts underlying the eTumorMetastasis pave the way for the application of genome sequencing in predictions for other complex genetic diseases. ETumorMetastasis pseudocode and related data used in this study can be found in our Github directory (https://github.com/WangEdwinLab/eTumorMetastasis).
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.gpb.2020.06.009DOI Listing
February 2021

FcγRIIA expression accelerates nephritis and increases platelet activation in systemic lupus erythematosus.

Blood 2020 12;136(25):2933-2945

Centre de Recherche du Centre Hospitalier Universitaire de Québec-Université Laval, Québec, QC, Canada.

Systemic lupus erythematosus (SLE) is an autoimmune inflammatory disease characterized by deposits of immune complexes (ICs) in organs and tissues. The expression of FcγRIIA by human platelets, which is their unique receptor for immunoglobulin G antibodies, positions them to ideally respond to circulating ICs. Whereas chronic platelet activation and thrombosis are well-recognized features of human SLE, the exact mechanisms underlying platelet activation in SLE remain unknown. Here, we evaluated the involvement of FcγRIIA in the course of SLE and platelet activation. In patients with SLE, levels of ICs are associated with platelet activation. Because FcγRIIA is absent in mice, and murine platelets do not respond to ICs in any existing mouse model of SLE, we introduced the FcγRIIA (FCGR2A) transgene into the NZB/NZWF1 mouse model of SLE. In mice, FcγRIIA expression by bone marrow cells severely aggravated lupus nephritis and accelerated death. Lupus onset initiated major changes to the platelet transcriptome, both in FcγRIIA-expressing and nonexpressing mice, but enrichment for type I interferon response gene changes was specifically observed in the FcγRIIA mice. Moreover, circulating platelets were degranulated and were found to interact with neutrophils in FcγRIIA-expressing lupus mice. FcγRIIA expression in lupus mice also led to thrombosis in lungs and kidneys. The model recapitulates hallmarks of human SLE and can be used to identify contributions of different cellular lineages in the manifestations of SLE. The study further reveals a role for FcγRIIA in nephritis and in platelet activation in SLE.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1182/blood.2020004974DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7751357PMC
December 2020

Identification of a Transcriptomic Prognostic Signature by Machine Learning Using a Combination of Small Cohorts of Prostate Cancer.

Front Genet 2020 25;11:550894. Epub 2020 Nov 25.

Centre de Recherche du CHU de Québec - Université Laval, Québec, QC, Canada.

Determining which treatment to provide to men with prostate cancer (PCa) is a major challenge for clinicians. Currently, the clinical risk-stratification for PCa is based on clinico-pathological variables such as Gleason grade, stage and prostate specific antigen (PSA) levels. But transcriptomic data have the potential to enable the development of more precise approaches to predict evolution of the disease. However, high quality RNA sequencing (RNA-seq) datasets along with clinical data with long follow-up allowing discovery of biochemical recurrence (BCR) biomarkers are small and rare. In this study, we propose a machine learning approach that is robust to batch effect and enables the discovery of highly predictive signatures despite using small datasets. Gene expression data were extracted from three RNA-Seq datasets cumulating a total of 171 PCa patients. Data were re-analyzed using a unique pipeline to ensure uniformity. Using a machine learning approach, a total of 14 classifiers were tested with various parameters to identify the best model and gene signature to predict BCR. Using a random forest model, we have identified a signature composed of only three genes (JUN, HES4, PPDPF) predicting BCR with better accuracy [74.2%, balanced error rate (BER) = 27%] than the clinico-pathological variables (69.2%, BER = 32%) currently in use to predict PCa evolution. This score is in the range of the studies that predicted BCR in single-cohort with a higher number of patients. We showed that it is possible to merge and analyze different small and heterogeneous datasets altogether to obtain a better signature than if they were analyzed individually, thus reducing the need for very large cohorts. This study demonstrates the feasibility to regroup different small datasets in one larger to identify a predictive genomic signature that would benefit PCa patients.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3389/fgene.2020.550894DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7723980PMC
November 2020

Immune-focused multi-omics analysis of prostate cancer: leukocyte Ig-Like receptors are associated with disease progression.

Oncoimmunology 2020 12 1;9(1):1851950. Epub 2020 Dec 1.

Computational Biology Laboratory, CHU de Québec - Université Laval Research Center, Québec City, QC, Canada.

Prostate cancer (PCa) immunotherapy has shown limited efficacy so far, even in advanced-stage cancers. The success rate of PCa immunotherapy might be improved by approaches more adapted to the immunobiology of the disease. The objective of this study was to perform a multi-omics analysis to identify immune genes associated with PCa progression to better characterize PCa immunobiology and propose new immunotherapeutic targets. mRNA, miRNA, methylation, copy number aberration, and single nucleotide variant datasets from The Cancer Genome Atlas PRAD cohort were analyzed after filtering for genes associated with immunity. Sparse partial least squares-discriminant analyses were performed to identify features associated with biochemical recurrence (BCR) in each type of omics data. Selected features predicted BCR with a balanced error rate (BER) of 0.20 to 0.51 in single-omics and of 0.05 in multi-omics analyses. Amongst features associated with BCR were genes from the Immunoglobulin Ig-like Receptor (LILR) family which are immune checkpoints with immunotherapeutic potential. Using Multivariate INTegrative (MINT) analysis, the association of five genes with BCR was quantified in a combination of three RNA-seq datasets and confirmed with Kaplan-Meier analysis in both these and in an independent RNA-seq dataset. Finally, immunohistochemistry showed that a high number of LILRB1 positive cells within the tumors predicted long-term adverse outcomes. Thus, tumors characterized by abnormal expression of genes have an elevated risk of recurring after definitive local therapy. The immunotherapeutic potential of these regulators to stimulate the immune response against PCa should be evaluated in pre-clinical models.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1080/2162402X.2020.1851950DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7714461PMC
December 2020

Omega-3 Eicosapentaenoic Acid Reduces Prostate Tumor Vascularity.

Mol Cancer Res 2021 03 1;19(3):516-527. Epub 2020 Dec 1.

Laboratoire d'Uro-Oncologie Expérimentale, Oncology Axis, Centre de recherche du CHU de Québec-Université Laval, Québec, Québec, Canada.

The impact of omega (ω)-3 fatty acids on prostate cancer is controversial in epidemiological studies but experimental studies suggest a protective effect. However, little is known about the mechanism of action. Here, we studied the effects of purified fatty acid molecules on prostate tumor progression using the TRAMP-C2 syngeneic immunocompetent mouse model. Compared with ω-6 or ω-9-supplemented animals, we observed that late-stage prostate tumor growth was reduced with a monoacylglyceride (MAG)-conjugated form of eicosapentaenoic acid (EPA) supplementation, whereas docosahexanenoic acid (DHA) caused an early reduction. MAG-EPA significantly decreased tumor blood vessel diameter ( < 0.001). RNA sequencing analysis revealed that MAG-EPA downregulated angiogenesis- and vascular-related pathways in tumors. We also observed this tissue vascular phenotype in a clinical trial testing MAG-EPA versus a high oleic sunflower oil placebo. Using anti-CD31 IHC, we observed that MAG-EPA reduced blood vessel diameter in prostate tumor tissue ( = 0.03) but not in normal adjacent tissue. Finally, testing autocrine and paracrine effects in an avascular tumor spheroid growth assay, both exogenous MAG-EPA and endogenous ω3 reduced VEGF secretion and endothelial cell tube formation and blocked tumor spheroid growth, suggesting that ω3 molecules can directly hinder prostate cancer cell growth. Altogether, our results suggest that fatty acids regulate prostate cancer growth and that a tumor-specific microenvironment is required for the anti-vascular effect of MAG-EPA in patients with prostate cancer. IMPLICATIONS: Increasing the amount of ingested EPA omega-3 subtype for patients with prostate cancer might help to reduce prostate tumor progression by reducing tumor vascularization.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1158/1541-7786.MCR-20-0316DOI Listing
March 2021

Cumulative learning enables convolutional neural network representations for small mass spectrometry data classification.

Nat Commun 2020 11 5;11(1):5595. Epub 2020 Nov 5.

Computational Biology Laboratory, CHU de Québec - Université Laval Research Center, Québec City, Québec, Canada., Québec City, QC, Canada.

Rapid and accurate clinical diagnosis remains challenging. A component of diagnosis tool development is the design of effective classification models with Mass spectrometry (MS) data. Some Machine Learning approaches have been investigated but these models require time-consuming preprocessing steps to remove artifacts, making them unsuitable for rapid analysis. Convolutional Neural Networks (CNNs) have been found to perform well under such circumstances since they can learn representations from raw data. However, their effectiveness decreases when the number of available training samples is small, which is a common situation in medicine. In this work, we investigate transfer learning on 1D-CNNs, then we develop a cumulative learning method when transfer learning is not powerful enough. We propose to train the same model through several classification tasks over various small datasets to accumulate knowledge in the resulting representation. By using rat brain as the initial training dataset, a cumulative learning approach can have a classification accuracy exceeding 98% for 1D clinical MS-data. We show the use of cumulative learning using datasets generated in different biological contexts, on different organisms, and acquired by different instruments. Here we show a promising strategy for improving MS data classification accuracy when only small numbers of samples are available.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1038/s41467-020-19354-zDOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7644674PMC
November 2020

Statistical and Machine-Learning Analyses in Nutritional Genomics Studies.

Nutrients 2020 Oct 14;12(10). Epub 2020 Oct 14.

Endocrinology and Nephrology Unit, CHU de Québec-Laval University Research Center, Quebec (PQ), QC G1V 4G2, Canada.

Nutritional compounds may have an influence on different OMICs levels, including genomics, epigenomics, transcriptomics, proteomics, metabolomics, and metagenomics. The integration of OMICs data is challenging but may provide new knowledge to explain the mechanisms involved in the metabolism of nutrients and diseases. Traditional statistical analyses play an important role in description and data association; however, these statistical procedures are not sufficiently enough powered to interpret the large integrated multiple OMICs (multi-OMICS) datasets. Machine learning (ML) approaches can play a major role in the interpretation of multi-OMICS in nutrition research. Specifically, ML can be used for data mining, sample clustering, and classification to produce predictive models and algorithms for integration of multi-OMICs in response to dietary intake. The objective of this review was to investigate the strategies used for the analysis of multi-OMICs data in nutrition studies. Sixteen recent studies aimed to understand the association between dietary intake and multi-OMICs data are summarized. Multivariate analysis in multi-OMICs nutrition studies is used more commonly for analyses. Overall, as nutrition research incorporated multi-OMICs data, the use of novel approaches of analysis such as ML needs to complement the traditional statistical analyses to fully explain the impact of nutrition on health and disease.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3390/nu12103140DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7602401PMC
October 2020

Identification of the most important features of knee osteoarthritis structural progressors using machine learning methods.

Ther Adv Musculoskelet Dis 2020 13;12:1759720X20933468. Epub 2020 Aug 13.

Osteoarthritis Research Unit, University of Montreal Hospital Research Centre (CRCHUM), 900 Saint-Denis, Suite R11.412, Montreal, Quebec H2X 0A9, Canada.

Objectives: The aim was to identify the most important features of structural knee osteoarthritis (OA) progressors and classification using machine learning methods.

Methods: Participants, features and outcomes were from the Osteoarthritis Initiative. Features were from baseline (1107), including articular knee tissues (135) assessed by quantitative magnetic resonance imaging (MRI). OA progressors were ascertained by four outcomes: cartilage volume loss in medial plateau at 48 and 96 months (Prop_CV_48M, 96M), Kellgren-Lawrence (KL) grade ⩾ 2 and medial joint space narrowing (JSN) ⩾ 1 at 48 months. Six feature selection models were used to identify the common features in each outcome. Six classification methods were applied to measure the accuracy of the selected features in classifying the subjects into progressors and non-progressors. Classification of the best features was done using an automatic machine learning interface and the area under the curve (AUC). To prioritize the top five features, sparse partial least square (sPLS) method was used.

Results: For the classification of the best common features in each outcome, Multi-Layer Perceptron (MLP) achieved the highest AUC in Prop_CV_96M, KL and JSN (0.80, 0.88, 0.95), and Gradient Boosting Machine for Prop_CV_48M (0.70). sPLS showed the baseline top five features to predict knee OA progressors are the joint space width, mean cartilage thickness of the medial tibial plateau and sub-regions and JSN.

Conclusion: In this comprehensive study using a large number of features ( = 1107) and MRI outcomes in addition to radiological outcomes, we identified the best features and classification methods for knee OA structural progressors. Data revealed baseline X-ray and MRI-based features could predict early OA knee progressors and that MLP is the best classification method.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1177/1759720X20933468DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7427139PMC
August 2020
-->