SNP characteristics predict replication success in association studies

Screening of Pathogenic Missense Single Nucleotide Variants From LHPP Gene Associated With the Hepatocellular Carcinoma: An In silico Approach

LHPP gene encodes a phospholysine phosphohistidine inorganic pyrophosphate phosphatase, which functions as a tumor-suppressor protein. The tumor suppression by this protein has been confirmed in various cancers, including hepatocellular carcinoma (HCC). LHPP downregulation promotes cell growth and proliferation by modulating the PI3K/AKT signaling pathway. This study identifies potentially deleterious missense single nucleotide variants (SNVs) associated with the LHPP gene using multiple computational tools based on different algorithms. A total of 4 destabilizing mutants are identified as L22P, I212T, G227R, and G236R, from the conserved region of the phosphatase. The 3-dimensional (3D) modeling and structural comparison of variants with the native protein reveals significant structural and conformational variations after mutations, suggesting disruption in the function of phospholysine phosphohistidine inorganic pyrophosphate phosphatase. The identified mutations might, therefore, participate in the cause of HCC.

SNP characteristics and validation success in genome wide association studies

Genome wide association studies (GWASs) have identified tens of thousands of single nucleotide polymorphisms (SNPs) associated with human diseases and characteristics. A significant fraction of GWAS findings can be false positives. The gold standard for true positives is an independent validation. The goal of this study was to identify SNP features associated with validation success. Summary statistics from the Catalog of Published GWASs were used in the analysis. Since our goal was an analysis of reproducibility, we focused on the diseases/phenotypes targeted by at least 10 GWASs. GWASs were arranged in discovery-validation pairs based on the time of publication, with the discovery GWAS published before validation. We used four definitions of the validation success that differ by stringency. Associations of SNP features with validation success were consistent across the definitions. The strongest predictor of SNP validation was the level of statistical significance in the discovery GWAS. The magnitude of the effect size was associated with validation success in a non-linear manner. SNPs with risk allele frequencies in the range 30-70% showed a higher validation success rate compared to rarer or more common SNPs. Missense, 5'UTR, stop gained, and SNPs located in transcription factor binding sites had a higher validation success rate compared to intergenic, intronic and synonymous SNPs. There was a positive association between validation success and the level of evolutionary conservation of the sites. In addition, validation success was higher when discovery and validation GWASs targeted the same ethnicity. All predictors of validation success remained significant in a multivariate logistic regression model indicating their independent contribution. To conclude, we identified SNP features predicting validation success of GWAS hits. These features can be used to select SNPs for validation and downstream functional studies.

Reproducibility in the UK biobank of genome-wide significant signals discovered in earlier genome-wide association studies

With the establishment of large biobanks, discovery of single nucleotide variants (SNVs, also known as single nucleotide polymorphisms (SNVs)) associated with various phenotypes has accelerated. An open question is whether genome-wide significant SNVs identified in earlier genome-wide association studies (GWAS) are replicated in later GWAS conducted in biobanks. To address this, we examined a publicly available GWAS database and identified two, independent GWAS on the same phenotype (an earlier, “discovery” GWAS and a later, “replication” GWAS done in the UK biobank). The analysis evaluated 136,318,924 SNVs (of which 6289 reached P < 5e−8 in the discovery GWAS) from 4,397,962 participants across nine phenotypes. The overall replication rate was 85.0%; although lower for binary than quantitative phenotypes (58.1% versus 94.8% respectively). There was a 18.0% decrease in SNV effect size for binary phenotypes, but a 12.0% increase for quantitative phenotypes. Using the discovery SNV effect size, phenotype trait (binary or quantitative), and discovery P value, we built and validated a model that predicted SNV replication with area under the Receiver Operator Curve = 0.90. While non-replication may reflect lack of power rather than genuine false-positives, these results provide insights about which discovered associations are likely to be replicated across subsequent GWAS.

Find the Needle in the Haystack, Then Find It Again: Replication and Validation in the 'Omics Era

Advancements in high-throughput technologies have made it feasible to study thousands of biological pathways simultaneously for a holistic assessment of health and disease risk via ‘omics platforms. A major challenge in ‘omics research revolves around the reproducibility of findings—a feat that hinges upon balancing false-positive associations with generalizability. Given the foundational role of reproducibility in scientific inference, replication and validation of ‘omics findings are cornerstones of this effort. In this narrative review, we define key terms relevant to replication and validation, present issues surrounding each concept with historical and contemporary examples from genomics (the most well-established and upstream ‘omics), discuss special issues and unique considerations for replication and validation in metabolomics (an emerging field and most downstream ‘omics for which best practices remain yet to be established), and make suggestions for future research leveraging multiple ‘omics datasets.

The Effect Sizes of PPARγ rs1801282, FTO rs9939609, and MC4R rs2229616 Variants on Type 2 Diabetes Mellitus Risk among the Western Saudi Population: A Cross-Sectional Prospective Study

Type 2 diabetes mellitus (T2DM) is a common polygenic disease with associated comorbidities. Obesity is a major risk factor for the development of T2DM. The aim of this study is to determine the allele and genotype frequency of peroxisome proliferator-activated receptor-γ (PPARγ) rs1801282, fat mass and obesity-associated protein (FTO) rs9939609, and melanocortin 4 receptor (MC4R) rs2229616 polymorphisms and their association with risk of T2DM in the western Saudi population as mediators of adiposity phenotypes. In a cross-sectional prospective study, genomic DNA from control and T2DM patients were isolated and genotyped for these single-nucleotide polymorphisms. There was a significant association of the MC4R rs2229616 variant with T2DM, but no association with T2DM was detected with PPARγ rs1801282 or FTO rs9939609. The combination of C/C for PPARγ rs1801282, A/A for FTO rs9939609, and C/C for MC4R rs2229616 increased the risk of T2DM by 1.82. The A/T genotype for FTO rs9939609 was predicted to decrease the risk of T2DM when combined with C/C for PPARγ rs1801282 and C/C for MC4R rs2229616 or C/C for PPARγ rs1801282 and C/T MC4R rs2229616. In conclusion, our study showed the risk of the assessed variants for the development of T2DM in the Saudi population.

SNP eQTL status and eQTL density in the adjacent region of the SNP are associated with its statistical significance in GWA studies

Background

Over the relatively short history of Genome Wide Association Studies (GWASs), hundreds of GWASs have been published and thousands of disease risk-associated SNPs have been identified. Summary statistics from the conducted GWASs are often available and can be used to identify SNP features associated with the level of GWAS statistical significance. Those features could be used to select SNPs from gray zones (SNPs that are nominally significant but do not reach the genome-wide level of significance) for targeted analyses.

Methods

We used summary statistics from recently published breast and lung cancer and scleroderma GWASs to explore the association between the level of the GWAS statistical significance and the expression quantitative trait loci (eQTL) status of the SNP. Data from the Genotype-Tissue Expression Project (GTEx) were used to identify eQTL SNPs.

Results

We found that SNPs reported as eQTLs were more significant in GWAS (higher -log₁₀p) regardless of the tissue specificity of the eQTL. Pan-tissue eQTLs (those reported as eQTLs in multiple tissues) tended to be more significant in the GWAS compared to those reported as eQTL in only one tissue type. eQTL density in the ±5 kb adjacent region of a given SNP was also positively associated with the level of GWAS statistical significance regardless of the eQTL status of the SNP. We found that SNPs located in the regions of high eQTL density were more likely to be located in regulatory elements (transcription factor or miRNA binding sites).

When SNPs were stratified by the level of statistical significance, the proportion of eQTLs was positively associated with the mean level of statistical significance in the group. The association curve reaches a plateau around -log₁₀p ≈ 5. The observed associations suggest that quasi-significant SNPs (10^− 5 < p < 5 × 10^− 8) and SNPs at the genome wide level of statistical significance (p < 5 × 10^− 8) may have a similar proportions of risk associated SNPs.

Conclusions

The results of this study indicate that the SNP’s eQTL status, as well as eQTL density in the adjacent region are positively associated with the level of statistical significance of the SNP in GWAS.

Untouchable genes in the human genome: Identifying ideal targets for cancer treatment

BACKGROUND

Usually, genes with a higher-than-expected number of somatic mutations in tumor samples are assumed to be cancer related. We identified genes with a fewer-than-expected number of somatic mutations - "untouchable genes".

METHODS

To predict the expected number of somatic mutations, we used a linear regression model with the number of mutations in the gene as an outcome, and gene characteristics, including gene size, nucleotide composition, level of evolutionary conservation, expression level and others, as predictors. Analysis of residuals from the regression model was used to compare the observed and predicted number of mutations.

RESULTS

We have identified 19 genes with a less-than-expected number of loss-off-function (nonsense, frameshift or pathogenic missense) mutations - i.e., untouchable genes. The number of silent or neutral missense mutations in untouchable genes was equal or higher than the expected number. Many mucins, including MUC16, MUC17, MUC6, MUC5AC, MUC5B, and MUC12, are untouchable. We hypothesized that untouchable mucins help tumor cells to avoid immune response by providing a protective coat that prevents direct contact between effector immune cells, e.g., cytotoxic T-cells, and tumor cells. Survival analysis of available TCGA data demonstrated that overall survival of patients with low (below the median) expression of untouchable mucins was better compared to patients with high expression of untouchable mucins. Aside from mucins, we have identified a number of other untouchable genes.

CONCLUSIONS

Untouchable genes may be ideal targets for cancer treatment since suppression of untouchable genes is expected to inhibit survival of tumor cells.

Gene characteristics predicting missense, nonsense and frameshift mutations in tumor samples

BACKGROUND

Because driver mutations provide selective advantage to the mutant clone, they tend to occur at a higher frequency in tumor samples compared to selectively neutral (passenger) mutations. However, mutation frequency alone is insufficient to identify cancer genes because mutability is influenced by many gene characteristics, such as size, nucleotide composition, etc. The goal of this study was to identify gene characteristics associated with the frequency of somatic mutations in the gene in tumor samples.

RESULTS

We used data on somatic mutations detected by genome wide screens from the Catalog of Somatic Mutations in Cancer (COSMIC). Gene size, nucleotide composition, expression level of the gene, relative replication time in the cell cycle, level of evolutionary conservation and other gene characteristics (totaling 11) were used as predictors of the number of somatic mutations. We applied stepwise multiple linear regression to predict the number of mutations per gene. Because missense, nonsense, and frameshift mutations are associated with different sets of gene characteristics, they were modeled separately. Gene characteristics explain 88% of the variation in the number of missense, 40% of nonsense, and 23% of frameshift mutations. Comparisons of the observed and expected numbers of mutations identified genes with a higher than expected number of mutations- positive outliers. Many of these are known driver genes. A number of novel candidate driver genes was also identified.

CONCLUSIONS

By comparing the observed and predicted number of mutations in a gene, we have identified known cancer-associated genes as well as 111 novel cancer associated genes. We also showed that adding the number of silent mutations per gene reported by genome/exome wide screens across all cancer type (COSMIC data) as a predictor substantially exceeds predicting accuracy of the most popular cancer gene predicting tool - MutsigCV.

Genome-wide association analysis of common genetic variants of resistant hypertension

Resistant hypertension (RHTN), defined as uncontrolled blood pressure (BP) ≥ 140/90 using three or more drugs or controlled BP (<140/90) using four or more drugs, is associated with adverse outcomes, including decline in kidney function. We conducted a genome-wide association analysis in 1194 White and Hispanic participants with hypertension and coronary artery disease from the INternational VErapamil-SR Trandolapril STudy-GENEtic Substudy (INVEST-GENES). Top variants associated with RHTN at p < 10^-4 were tested for replication in 585 White and Hispanic participants with hypertension and subcortical strokes from the Secondary Prevention of Subcortical Strokes GENEtic Substudy (SPS3-GENES). A genetic risk score for RHTN was created by summing the risk alleles of replicated RHTN signals. rs11749255 in MSX2 was associated with RHTN in INVEST (odds ratio (OR) (95% CI) = 1.50 (1.2-1.8), p = 7.3 × 10^-5) and replicated in SPS3 (OR = 2.0 (1.4-2.8), p = 4.3 × 10^-5), with genome-wide significance in meta-analysis (OR = 1.60 (1.3-1.9), p = 3.8 × 10^-8). Other replicated signals were in IFLTD1 and PTPRD. IFLTD1 rs6487504 was associated with RHTN in INVEST (OR = 1.90 (1.4-2.5), p = 1.1 × 10^-5) and SPS3 (OR = 1.70 (1.2-2.5), p = 4 × 10^-3). PTPRD rs324498, a previously reported RHTN signal, was among the top signals in INVEST (OR = 1.60 (1.3-2.0), p = 3.4 × 10^-5) and replicated in SPS3 (OR = 1.60 (1.1-2.4), one-sided p = 0.005). Participants with the highest number of risk alleles were at increased risk of RHTN compared to participants with a lower number (p-trend = 1.8 × 10^-15). Overall, we identified and replicated associations with RHTN in the MSX2, IFLTD1, and PTPRD regions, and combined these associations to create a genetic risk score.

Validation study of genetic biomarkers of response to TNF inhibitors in rheumatoid arthritis

Genetic biomarkers are sought to personalize treatment of patients with rheumatoid arthritis (RA), given their variable response to TNF inhibitors (TNFi). However, no genetic biomaker is yet sufficiently validated. Here, we report a validation study of 18 previously reported genetic biomarkers, including 11 from GWAS of response to TNFi. The validation was attempted in 581 patients with RA that had not been treated with biologic antirheumatic drugs previously. Their response to TNFi was evaluated at 3, 6 and 12 months in two ways: change in the DAS28 measure of disease activity, and according to the EULAR criteria for response to antirheumatic drugs. Association of these parameters with the genotypes, obtained by PCR amplification followed by single-base extension, was tested with regression analysis. These analyses were adjusted for baseline DAS28, sex, and the specific TNFi. However, none of the proposed biomarkers was validated, as none showed association with response to TNFi in our study, even at the time of assessment and with the outcome that showed the most significant result in previous studies. These negative results are notable because this was the first independent validation study for 12 of the biomarkers, and because they indicate that prudence is needed in the interpretation of the proposed biomarkers of response to TNFi even when they are supported by very low p values. The results also emphasize the requirement of independent replication for validation, and the need to search protocols that could increase reproducibility of the biomarkers of response to TNFi.

Association of Matrilin-3 Gene Polymorphism with Temporomandibular Joint Internal Derangement

AIMS

Temporomandibular joint internal derangement (TMJ ID) is a multifactorial complex disease characterised by articular disc degeneration. Matrilin-3 is a cartilage and bone-specific adaptor protein, and amino-acid substitutions in the protein are associated with skeletal diseases and joint disorders. We aimed to detect the variants of Matrilin-3 gene (MATN3) in a TMJ ID case-control group and to investigate the risk association of the detected variants with TMJ ID.

MATERIALS AND METHODS

A case control study was conducted consisting of 57 unrelated TMJ ID patients (32.7 ± 8.2) and 96 unrelated healthy controls (26.63 ± 3.05) without TMJ ID to look for associations with variants of the MATN3 gene. DNA from individual subjects was extracted and each of the eight exons was amplified by polymerase chain reaction using and analyzed by single-strand conformational polymorphism (SSCP) analysis. SSCP variants were subjected to DNA sequence analysis, which yielded band pattern variations in exon 2 of the gene. We further analyzed exon 2 by DNA sequencing to determine the sequence of these variants.

RESULTS

We identified SSCP band patterns variants in exon 2 of the MATN3 gene which upon sequencing revealed a single C to T transition mutation (rs28598872) c.447 C>T (g.11608 C>T). This polymorphism is predicted to result in a synonymous mutation (pAla149 = ). The TT and CT genotypes were more prevalent than the CC genotype in TMJ ID patients compared to the control group with a risk factor of 2.12 (confidence intervals [CI] :0.88-5.08) and 2.0 (CI:0.726-5.508). In addition, TMJ ID patients were divided into two groups as anterior disc displacement with reduction (ADDWR) and anterior disc displacement without reduction (ADDWOR) and compared with the controls. The TT and CT genotypes were more prevalent than the CC genotype in ADDWR patients compared to the control group with a risk factor of 3.85 (CI:0.927-16.048) and 3.75 (1.02-13.786), respectively. We found that, among ADDWR patients, the T allele is a risk factor both in homozygous and heterozygous carriers (p < 0.052, p < 0.036).

CONCLUSION

The results of the study indicate a potential role for the MATN3 rs28598872 polymorphism in the pathogenesis of TMJ ID.

The Use of Multiplicity Corrections, Order Statistics and Generalized Family-Wise Statistics with Application to Genome-Wide Studies

The most important decision faced by large-scale studies, such as those presently encountered in human genetics, is to distinguish between those tests that are true positives from those that are not. In the context of genetics, this entails the determination of genetic markers that actually underlie medically-relevant phenotypes from a vast number of makers typically interrogated in genome-wide studies. A critical part of these decisions relies on the appropriate statistical assessment of data obtained from tests across numerous markers. Several methods have been developed to aid with such analyses, with family-wise approaches, such as the Bonferroni and Dunn-Šidàk corrections, being popular. Conditions that motivate the use of family-wise corrections are explored. Although simple to implement, one major limitation of these approaches is that they assume that p-values are i.i.d. uniformly distributed under the null hypothesis. However, several factors may violate this assumption in genome-wide studies including effects from confounding by population stratification, the presence of related individuals, the correlational structure among genetic markers, and the use of limiting distributions for test statistics. Even after adjustment for such effects, the distribution of p-values can substantially depart from a uniform distribution under the null hypothesis. In this work, I present a decision theory for the use of family-wise corrections for multiplicity and a generalization of the Dunn-Šidàk correction that relaxes the assumption of uniformly-distributed null p-values. The independence assumption is also relaxed and handled through calculating the effective number of independent tests. I also explicitly show the relationship between order statistics and family-wise correction procedures. This generalization may be applicable to multiplicity problems outside of genomics.

Machine learning and data mining in complex genomic data--a review on the lessons learned in Genetic Analysis Workshop 19

In the analysis of current genomic data, application of machine learning and data mining techniques has become more attractive given the rising complexity of the projects. As part of the Genetic Analysis Workshop 19, approaches from this domain were explored, mostly motivated from two starting points. First, assuming an underlying structure in the genomic data, data mining might identify this and thus improve downstream association analyses. Second, computational methods for machine learning need to be developed further to efficiently deal with the current wealth of data.In the course of discussing results and experiences from the machine learning and data mining approaches, six common messages were extracted. These depict the current state of these approaches in the application to complex genomic data. Although some challenges remain for future studies, important forward steps were taken in the integration of different data types and the evaluation of the evidence. Mining the data for underlying genetic or phenotypic structure and using this information in subsequent analyses proved to be extremely helpful and is likely to become of even greater use with more complex data sets.

Reconciling Differences in Pool-GWAS Between Populations: A Case Study of Female Abdominal Pigmentation in Drosophila melanogaster

The degree of concordance between populations in the genetic architecture of a given trait is an important issue in medical and evolutionary genetics. Here, we address this problem, using a replicated pooled genome-wide association study approach (Pool-GWAS) to compare the genetic basis of variation in abdominal pigmentation in female European and South African Drosophila melanogaster. We find that, in both the European and the South African flies, variants near the tan and bric-à-brac 1 (bab1) genes are most strongly associated with pigmentation. However, the relative contribution of these loci differs: in the European populations, tan outranks bab1, while the converse is true for the South African flies. Using simulations, we show that this result can be explained parsimoniously, without invoking different causal variants between the populations, by a combination of frequency differences between the two populations and dominance for the causal alleles at the bab1 locus. Our results demonstrate the power of cost-effective, replicated Pool-GWAS to shed light on differences in the genetic architecture of a given trait between populations.

Prediction of the gene expression in normal lung tissue by the gene expression in blood

BACKGROUND

Comparative analysis of gene expression in human tissues is important for understanding the molecular mechanisms underlying tissue-specific control of gene expression. It can also open an avenue for using gene expression in blood (which is the most easily accessible human tissue) to predict gene expression in other (less accessible) tissues, which would facilitate the development of novel gene expression based models for assessing disease risk and progression. Until recently, direct comparative analysis across different tissues was not possible due to the scarcity of paired tissue samples from the same individuals.

METHODS

In this study we used paired whole blood/lung gene expression data from the Genotype-Tissue Expression (GTEx) project. We built a generalized linear regression model for each gene using gene expression in lung as the outcome and gene expression in blood, age and gender as predictors.

RESULTS

For ~18 % of the genes, gene expression in blood was a significant predictor of gene expression in lung. We found that the number of single nucleotide polymorphisms (SNPs) influencing expression of a given gene in either blood or lung, also known as the number of quantitative trait loci (eQTLs), was positively associated with efficacy of blood-based prediction of that gene's expression in lung. This association was strongest for shared eQTLs: those influencing gene expression in both blood and lung.

CONCLUSIONS

In conclusion, for a considerable number of human genes, their expression levels in lung can be predicted using observable gene expression in blood. An abundance of shared eQTLs may explain the strong blood/lung correlations in the gene expression.