WISH-R– a fast and efficient tool for construction of epistatic networks for complex traits and diseases

Carmelo, Victor Adriano Okstoft; Kogelman, Lisette J. A.; Madsen, Majbritt Busk; Kadarmideen, Haja N.

doi:10.1186/s12859-018-2291-2

Cited by 12 publications

(16 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The WGCNA methods have been successfully applied to gene expression data from microarrays 60 and RNA sequencing platforms in animal sciences 18 . Recently this methodology was applied on genome-wide genotype data as well 61 . Hereby, we extended this methodology to build networks using metabolomics data.…”

Section: Metabolite Network Analysis Network Analysis Was Performed mentioning

confidence: 99%

Metabolomic networks and pathways associated with feed efficiency and related-traits in Duroc and Landrace pigs

Carmelo

Banerjee

Diniz

et al. 2020

Sci Rep

Self Cite

View full text Add to dashboard Cite

Improving feed efficiency (FE) is a major goal of pig breeding, reducing production costs and providing sustainability to the pig industry. Reliable predictors for fe could assist pig producers. We carried out untargeted blood metabolite profiling in uncastrated males from Danbred Duroc (n = 59) and Danbred Landrace (n = 50) pigs at the beginning and end of a FE testing phase to identify biomarkers and biological processes underlying fe and related traits. By applying linear modeling and clustering analyses coupled with WGCNA framework, we identified 102 and 73 relevant metabolites in Duroc and Landrace based on two sampling time points. Among them, choline and pyridoxamine were hub metabolites in Duroc in early testing phase, while, acetoacetate, cholesterol sulfate, xanthine, and deoxyuridine were identified in the end of testing. In Landrace, cholesterol sulfate, thiamine, L-methionine, chenodeoxycholate were identified at early testing phase, while, D-glutamate, pyridoxamine, deoxycytidine, and L-2-aminoadipate were found at the end of testing. Validation of these results in larger populations could establish fe prediction using metabolomics biomarkers. We conclude that it is possible to identify a link between blood metabolite profiles and FE. These results could lead to improved nutrient utilization, reduced production costs, and increased fe.With the expanding human population and requirement for nutrient-rich food, there is an increasing demand for improvement of meat production, but simultaneously, to decrease the input costs in terms of feed 1 . Thus, feed efficiency (FE) is the most important trait in commercial pig farming 2 as increasing the amount of meat produced per feed is beneficial both economically and environmentally. Thereby, improving FE is beneficial for producers and increases the sustainability of pork meat production. Fortunately, FE is a highly heritable trait in Danish pigs (ranging from 0.34 in Duroc to 0.40 in Landrace), thus suitable for the genetic selection of pigs with high breeding values in breeding programs aimed at improving this economically important phenotype 3 .Since FE cannot be measured directly, feed conversion ratio (FCR) and residual feed intake (RFI) have been used to evaluate the animal efficiency 4 . FCR determines the ratio of feed intake (FI) to output and found to correlate with growth rate and body weight 3,5 . RFI calculates the difference between the actual and expected FI 6 predicted based on production traits such as average daily gain (ADG) 7 . ADG is also considered important in commercial pig production as pigs with higher ADG can achieve a target market weight within a shorter period than those with lower ADG, thereby saving feeding costs 8 . Thus, selection for RFI has proved to be effective in improving the FE in pigs 3,9,10 . Selection for FCR will results in co-selection for other traits, such as body composition and ADG. In contrast, RFI selects for increased metabolic efficiency without the same side effects [11][12][13] . RFI and FCR are well c...

show abstract

Section: Metabolite Network Analysis Network Analysis Was Performed mentioning

confidence: 99%

Metabolomic networks and pathways associated with feed efficiency and related-traits in Duroc and Landrace pigs

Carmelo

Banerjee

Diniz

et al. 2020

Sci Rep

Self Cite

View full text Add to dashboard Cite

show abstract

“…We carried out a genome-wide association to investigate the epistatic interactions for feed efficiency traits in Duroc and Landrace pigs by applying the WISH-R package (Carmelo et al, 2018) based on WISH method (Kogelman and Kadarmideen, 2014). To identify breed-specific and group-specific SNP interactions, we selected the top 7,000 SNPs associated to FCR from each analysis, and then carried out the epistatic interaction calculation individually for each breed and each group (LFE, HFE).…”

Section: Genome-wide Pair-wise Interaction Analysismentioning

confidence: 99%

“…To identify breed-specific and group-specific SNP interactions, we selected the top 7,000 SNPs associated to FCR from each analysis, and then carried out the epistatic interaction calculation individually for each breed and each group (LFE, HFE). The selected SNPs were pruned for linkage disequilibrium (LD) using "LD_blocks" function, with a maximum block size of 1000, and threshold = 0.9 from WISH-R package (Carmelo et al, 2018). Further, the "epistatic.correlation" function, considering the default parameters, was employed to calculate the epistatic interaction among the remaining SNPs (Kogelman and Kadarmideen, 2014;Carmelo et al, 2018).…”

Section: Genome-wide Pair-wise Interaction Analysismentioning

confidence: 99%

“…WISH (Weighted Interaction SNP Hub) network method (Kogelman and Kadarmideen, 2014) has been proposed to calculate genomewide epistatic interactions and to construct epistatic networks underlying complex phenotypes. Recently implemented in R, as WISH-R package (Carmelo et al, 2018), this method detects genome-wide interactions between single nucleotide polymorphism (SNP) and traits in high throughput genomic data (HTG). WISH-R calculates epistasis and constructs biological networks using Weighted Gene Co-expression Network Analysis (WGCNA) framework (Langfelder and Horvath, 2008).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Genome-Wide Epistatic Interaction Networks Affecting Feed Efficiency in Duroc and Landrace Pigs

2020

Self Cite

View full text Add to dashboard Cite

Interactions among genomic loci have often been overlooked in genome-wide association studies, revealing the combinatorial effects of variants on phenotype or disease manifestation. Unexplained genetic variance, interactions among causal genes of small effects, and biological pathways could be identified using a network biology approach. The main objective of this study was to determine the genome-wide epistatic variants affecting feed efficiency traits [feed conversion ratio (FCR) and residual feed intake (RFI)] based on weighted interaction SNP hub (WISH-R) method. Herein, we detected highly interconnected epistatic SNP modules, pathways, and potential biomarkers for the FCR and RFI in Duroc and Landrace purebreds considering the whole population, and separately for low and high feed efficient groups. Highly interacting SNP modules in Duroc (1,247 SNPs) and Landrace (1,215 SNPs) across the population and for low feed efficient (Duroc-80 SNPs, Landrace-146 SNPs) and high feed efficient group (Duroc-198 SNPs, Landrace-232 SNPs) for FCR and RFI were identified. Gene and pathway analyses identified ABL1, MAP3K4, MAP3K5, SEMA6A, KITLG, and KAT2B from chromosomes 1, 2, 5, and 13 underlying ErbB, Ras, Rap1, thyroid hormone, axon guidance pathways in Duroc. GABBR2, GNA12, and PRKCG genes from chromosomes 1, 3, and 6 pointed towards thyroid hormone, cGMP-PKG and cAMP pathways in Landrace. From Duroc low feed efficient group, the TPK1 gene was found involved with thiamine metabolism, whereas PARD6G, DLG2, CRB1 were involved with the hippo signaling pathway in high feed efficient group. PLOD1 and SETD7 genes were involved with lysine degradation in low feed efficient group in Landrace, while high feed efficient group pointed to genes underpinning valine, leucine, isoleucine degradation, and fatty acid elongation. Some SNPs and genes identified are known for their association with feed efficiency, others are novel and potentially provide new avenues for further research. Further validation of epistatic SNPs and genes identified here in a larger cohort would help to establish a framework for modelling epistatic variance in future methods of genomic prediction, increasing the accuracy of estimated genetic merit for FE and helping the pig breeding industry.

show abstract

“…This reduced the initial set of SNPs to a total of 27531. The next step performed was to remove groups of SNPs in high LD To do this, we used the LD_blocks function from the WISH-R R package (31), which was applied with an R 2 of 0.9. This grouped SNPs linearly across chromosomes into blocks based on a minimum pairwise R 2 value of 0.9 between all SNPs in a block.…”

Section: Gentoype Data and Filteringmentioning

confidence: 99%

Genetic variations (eQTLs) in muscle transcriptome and mitochondrial genes, and trans-eQTL molecular pathways in feed efficiency from Danish breeding pigs

Carmelo

Kadarmideen

2020

Preprint

Self Cite

View full text Add to dashboard Cite

12Feed efficiency (FE) is a key trait in pig production, as it has both high economic and 13 environmental impact. FE is a challenging phenotype to study, as it is complex and affected by 14 many factors, such as metabolism, growth and activity level. Furthermore, testing for FE is 15 expensive, as it requires costly equipment to measure feed intake of individual animals, making 16 FE biomarkers valuable. Therefore, there has been a desire to find single nucleotide 17 polymorphisms (SNPs) as biomarkers, to assist with improved selection and improve our 18 biological understanding of FE. We have done a cis-and trans-eQTL (expressed quantitative 19 trait loci) analysis, in a population of Danbred Durocs (N=11) and Danbred Landrace (N=27) 20 using both a linear and Anova model. We used bootstrapping and enrichment analysis to 21 validate and analyze our detected eQTLs. We identified 15 eQTLs with FDR < 0.01, affecting 22 several genes found in previous studies of commercial pig breeds. Examples include IFI6, 23 PRPF39, TMEM222, CSRNP1,PARK7 and MFF. The bootstrapping results showed statistically 24 significant enrichment of eQTLs with p-value < 0.01 (p-value < 2.2x0 -16 ) in both cis and trans-25 eQTLs. Based on this, enrichment analysis of top trans-eQTLs revealed high enrichment for 26 gene categories and gene ontologies associated with genomic context and expression regulation.27 This includes transcription factors (p-value=1.0x10 -13 ), DNA-binding (GO:0003677, p-28 value=8.9x10 -14 ), DNA-binding transcription factor activity (GO:0003700,) nucleus gene 29 (GO:0005634, p-value<2.2x10 -16 ), positive regulation of expression (GO:0010628), negative 30 regulation of expression (GO:0010629, p-value<2.2x10 -16 ). These results would be useful for 31 future genome assisted breeding of pigs to improve FE, and in the improved understanding of 32 the functional mechanism of trans-eQTLs. 33 34 42 variation based on pathway and functional knowledge of implicated genes. This can be done through 43 the identification of expressed quantitative loci (eQTL), mapping genetic variants that influence gene 44 expression patterns of genes in various tissues, originally termed as systems genetics (5, 6). The usage 45 of both the genetic and the transcriptomic information, combined with pathway and phenotype data 46 can be a powerful way of identifying biomarkers for traits of interest. There are however, several 47 challenges with eQTL analysis. Firstly, if one wanted to map all possible SNP-gene pairs in a modern 48data set, which typically has thousands of expressed genes and at the minimum several tens of 49 thousands of SNP, the total amount of tests will be at least in the order of 10 8 . This can pose 50 computational challenges, but even worse, multiple testing problems. This is especially relevant as a 51 cursory search of the Gene Expression Omnibus database (https://www.ncbi.nlm.nih.gov/geo/) for 52 RNA-seq studies reveals most studies having less than 100 samples. Therefore, it is important to have 53 strategies for these ...

show abstract

WISH-R– a fast and efficient tool for construction of epistatic networks for complex traits and diseases

Cited by 12 publications

References 23 publications

Metabolomic networks and pathways associated with feed efficiency and related-traits in Duroc and Landrace pigs

Metabolomic networks and pathways associated with feed efficiency and related-traits in Duroc and Landrace pigs

Genome-Wide Epistatic Interaction Networks Affecting Feed Efficiency in Duroc and Landrace Pigs

Genetic variations (eQTLs) in muscle transcriptome and mitochondrial genes, and trans-eQTL molecular pathways in feed efficiency from Danish breeding pigs

Contact Info

Product

Resources

About