diff --git a/404.html b/404.html index 27e7d16..e13cce6 100644 --- a/404.html +++ b/404.html @@ -6,12 +6,6 @@ Page not found (404) • mvMAPIT - - - - - - @@ -26,7 +20,7 @@ - +
- +
@@ -132,16 +126,16 @@

Page not found (404)

-

Site built with pkgdown 2.0.9.

+

Site built with pkgdown 2.1.1.

- - + + diff --git a/LICENSE.html b/LICENSE.html index 25bfcf5..c7c7d13 100644 --- a/LICENSE.html +++ b/LICENSE.html @@ -1,9 +1,9 @@ -GNU General Public License • mvMAPITGNU General Public License • mvMAPIT - +
- +
@@ -289,15 +289,15 @@

How to Apply These Terms

-

Site built with pkgdown 2.0.9.

+

Site built with pkgdown 2.1.1.

- - + + diff --git a/apple-touch-icon-120x120.png b/apple-touch-icon-120x120.png deleted file mode 100644 index dbdb73c..0000000 Binary files a/apple-touch-icon-120x120.png and /dev/null differ diff --git a/apple-touch-icon-152x152.png b/apple-touch-icon-152x152.png deleted file mode 100644 index 0f98c7c..0000000 Binary files a/apple-touch-icon-152x152.png and /dev/null differ diff --git a/apple-touch-icon-180x180.png b/apple-touch-icon-180x180.png deleted file mode 100644 index edf79f0..0000000 Binary files a/apple-touch-icon-180x180.png and /dev/null differ diff --git a/apple-touch-icon-60x60.png b/apple-touch-icon-60x60.png deleted file mode 100644 index 849dc90..0000000 Binary files a/apple-touch-icon-60x60.png and /dev/null differ diff --git a/apple-touch-icon-76x76.png b/apple-touch-icon-76x76.png deleted file mode 100644 index 6ff5f20..0000000 Binary files a/apple-touch-icon-76x76.png and /dev/null differ diff --git a/apple-touch-icon.png b/apple-touch-icon.png deleted file mode 100644 index 12d3b8f..0000000 Binary files a/apple-touch-icon.png and /dev/null differ diff --git a/articles/index.html b/articles/index.html index eb44d0d..b6eb7b2 100644 --- a/articles/index.html +++ b/articles/index.html @@ -1,9 +1,9 @@ -Articles • mvMAPITArticles • mvMAPIT - +
- +
@@ -109,15 +109,15 @@

Studies

-

Site built with pkgdown 2.0.9.

+

Site built with pkgdown 2.1.1.

- - + + diff --git a/articles/mvMAPIT.html b/articles/mvMAPIT.html index e59f25e..bcc125a 100644 --- a/articles/mvMAPIT.html +++ b/articles/mvMAPIT.html @@ -6,12 +6,6 @@ Illustrating multivariate MAPIT with Simulated Data • mvMAPIT - - - - - - @@ -28,7 +22,7 @@ - +
- +
@@ -115,9 +109,9 @@

Illustrating multivariate MAPIT with Simulated Data

Julian Stamp

-

2024-06-07

+

2024-10-21

- Source: vignettes/mvMAPIT.Rmd + Source: vignettes/mvMAPIT.Rmd
@@ -155,7 +149,7 @@

Running mvMAPIT

The R routine mvmapit can be run in multiple modes. By default it runs in a hybrid mode, performing tests both wtih a normal Z-test as well as the Davies method. The resulting p-values can be combined using functions provided by mvMAPIT, e.g. fishers_combined(), that work on the pvalues tibble that mvmapit returns.

-

NOTE: mvMAPIT takes the X matrix as \(p \times n\); not as \(n \times p\).

+

NOTE: mvMAPIT takes the X matrix as p×np \times n; not as n×pn \times p.

 mvmapit_hybrid <- mvmapit(
@@ -369,7 +363,7 @@
@@ -693,16 +684,16 @@

True epistataic structure

-

Site built with pkgdown 2.0.9.

+

Site built with pkgdown 2.1.1.

- - + + diff --git a/articles/study-compare-p-value-combine-methods.html b/articles/study-compare-p-value-combine-methods.html index 6a7cc1d..560eaa0 100644 --- a/articles/study-compare-p-value-combine-methods.html +++ b/articles/study-compare-p-value-combine-methods.html @@ -6,12 +6,6 @@ Empirical comparison of P-value combination methods in mvMAPIT • mvMAPIT - - - - - - @@ -28,7 +22,7 @@ - +
- +
@@ -114,7 +108,7 @@

Empirical comparison of P-value combination methods in mvMAPIT

- Source: vignettes/study-compare-p-value-combine-methods.Rmd + Source: vignettes/study-compare-p-value-combine-methods.Rmd
@@ -126,20 +120,20 @@

Empirical comparison of P-value combination methods in mvMAPIT library(GGally) library(tidyr) library(dplyr)

-

In Stamp et al. (2023)1 we discuss the meta analysis of the pariwise comparison \(P\)-values that mvMAPIT gives. We provide three methods to compute a per-variant combined \(P\)-value:

+

In Stamp et al. (2023)1 we discuss the meta analysis of the pariwise comparison PP-values that mvMAPIT gives. We provide three methods to compute a per-variant combined PP-value:

  1. Fisher’s Method2
    2. -
  2. The harmonic mean \(P\)3 +
  3. The harmonic mean PP3
  4. The Cauchy combination test4
-

The Fisher’s method requires independent \(P\)-values. The other two tests handle arbitrary covariances between the \(P\)-values. Here we show how these three methods compare empirically when applied to the same \(P\)-values.

+

The Fisher’s method requires independent PP-values. The other two tests handle arbitrary covariances between the PP-values. Here we show how these three methods compare empirically when applied to the same PP-values.

Generate data

-

Draw random uniform \(P\)-values from [0, 0.05].

+

Draw random uniform PP-values from [0, 0.05].

 n_variants <- 10000
 n_combine <- 3
@@ -150,9 +144,9 @@ 
Generate data)
-

Combine \(P\)-Values with all methods +

Combine PP-Values with all methods

-

Use the provided methods to compute the meta analysis \(P\)-values.

+

Use the provided methods to compute the meta analysis PP-values.

 cauchy <- cauchy_combined(pvalues) %>%
   rename(p_cauchy = p) %>%
@@ -179,7 +173,7 @@ 
Combine \(P
 

 
Plot the result
 

-
The figure shows the data distribution on the diagonal and paired 2D historgam plots for all combinations of \(P\)-values. The brighter yellows correspond to higher counts, the green is in the middle of the scale, and the darker blues correspond to the low values of the histogram.

+
The figure shows the data distribution on the diagonal and paired 2D historgam plots for all combinations of PP-values. The brighter yellows correspond to higher counts, the green is in the middle of the scale, and the darker blues correspond to the low values of the histogram.

 
 my_bin <- function(data, mapping) {
   ggplot(data = data, mapping = mapping) +
@@ -209,9 +203,7 @@ 
References
 
-        
-

+
@@ -224,16 +216,16 @@

References

-

Site built with pkgdown 2.0.9.

+

Site built with pkgdown 2.1.1.

- - + + diff --git a/articles/study-phillips-bnabs.html b/articles/study-phillips-bnabs.html index 20edd10..c3d92ad 100644 --- a/articles/study-phillips-bnabs.html +++ b/articles/study-phillips-bnabs.html @@ -6,12 +6,6 @@ Synergistic epistasis in binding affinity landscapes • mvMAPIT - - - - - - @@ -28,7 +22,7 @@ - +
- +
@@ -114,7 +108,7 @@

Synergistic epistasis in binding affinity landscapes

- Source: vignettes/study-phillips-bnabs.Rmd + Source: vignettes/study-phillips-bnabs.Rmd
@@ -125,11 +119,11 @@

Synergistic epistasis in binding affinity landscapes

library(mvMAPIT) library(ggplot2) library(dplyr)
-

This study uses the data of the binding affinites published by Phillips et al. (2021)\(^1\).

+

This study uses the data of the binding affinites published by Phillips et al. (2021)1^1.

The Data

-

The data is comprised of a nearly combinatorially complete library for two broadly neutralizing anti-influenza antibodies (bnAbs), CR6261 and CR9114. This dataset includes almost all combinations of one-off mutations that distinguish between germline and somatic sequences which total to 11 heavy-chain mutations for CR6261 and 16 heavy-chain mutations for CR9114. Theoretically, a combinatorially complete dataset for 11 and 16 mutations will have 2,048 and 65,536 samples, respectively. In this particular study, we have have access to \(N =\) 1,812 complete observations for CR6261 and \(N =\) 65,091 complete measurements for CR9114. The traits are binding affinities of the two antibodies to different influenza strains.

+

The data is comprised of a nearly combinatorially complete library for two broadly neutralizing anti-influenza antibodies (bnAbs), CR6261 and CR9114. This dataset includes almost all combinations of one-off mutations that distinguish between germline and somatic sequences which total to 11 heavy-chain mutations for CR6261 and 16 heavy-chain mutations for CR9114. Theoretically, a combinatorially complete dataset for 11 and 16 mutations will have 2,048 and 65,536 samples, respectively. In this particular study, we have have access to N=N = 1,812 complete observations for CR6261 and N=N = 65,091 complete measurements for CR9114. The traits are binding affinities of the two antibodies to different influenza strains.

Download the data for the broadly neutralizing antibodies here:

  • @@ -141,7 +135,7 @@

    The Data

    Multivariate MAPIT Analysis

    -

    For our analysis with mvMAPIT, residue sequence information was encoded as a binary matrix with the germline sequence residues marked by zeros and the somatic mutations represented as ones. As quantitative traits, Phillips et al. (2021) measure the binding affinity of the two antibodies to different influenza strains. Here, we assess the contribution of epistatic effects when binding to \(H_1\) and \(H_9\) for CR6261, and \(H_1\) and \(H_3\) for CR9114.

    +

    For our analysis with mvMAPIT, residue sequence information was encoded as a binary matrix with the germline sequence residues marked by zeros and the somatic mutations represented as ones. As quantitative traits, Phillips et al. (2021) measure the binding affinity of the two antibodies to different influenza strains. Here, we assess the contribution of epistatic effects when binding to H1H_1 and H9H_9 for CR6261, and H1H_1 and H3H_3 for CR9114.

    Load the data

    @@ -177,7 +171,7 @@

    Load the data

    Apply mvMAPIT

    -

    We apply the mvMAPIT framework to protein sequence data from Phillips et al (2021)\(^1\).

    +

    We apply the mvMAPIT framework to protein sequence data from Phillips et al (2021)1^1.

    Apply mvmapit() to the data by running the following.

     mvmapit_CR9114 <- mvmapit(
@@ -194,8 +188,8 @@ 
    Apply mvMAPIT
 
    Plot the mvMAPIT results
 
    
-
    The package contains a data object with the results. This data is part of the namespace and accessible as phillips_data. It contains tibbles with the main column for inference being the \(P\)-values.
    
-
    We report results after running mvMAPIT and combined \(P\)-values with Fisher’s method. We show Manhattan plots for \(P\)-values corresponding to the trait-specific marginal epistatic tests (i.e., the univariate MAPIT model), the covariance test, and the mvMAPIT approach. Here, green colored dots are positions that have significant marginal epistatic effects beyond a Bonferroni corrected threshold for multiple testing (\(P = 0.05/11 = 4.55\times 10^{-3}\) for CR6261 and \(P = 0.05/16 = 3.13\times 10^{-3}\) for CR9114, respectively).
    
+
    The package contains a data object with the results. This data is part of the namespace and accessible as phillips_data. It contains tibbles with the main column for inference being the PP-values.
    
+
    We report results after running mvMAPIT and combined PP-values with Fisher’s method. We show Manhattan plots for PP-values corresponding to the trait-specific marginal epistatic tests (i.e., the univariate MAPIT model), the covariance test, and the mvMAPIT approach. Here, green colored dots are positions that have significant marginal epistatic effects beyond a Bonferroni corrected threshold for multiple testing (P=0.05/11=4.55×10−3P = 0.05/11 = 4.55\times 10^{-3} for CR6261 and P=0.05/16=3.13×10−3P = 0.05/16 = 3.13\times 10^{-3} for CR9114, respectively).
    
 
     for_facetgrid_row <-
   as_labeller(c(
@@ -256,7 +250,7 @@ 
    CR6261
   theme(legend.position = "bottom")
 show(gg_CR6261)
    
 
    
-
    This plot illustrates interaction coefficients when assessing binding of CR6261with \(H_1\) (lower left triangle) and \(H_9\) (upper right triangle).
    
+
    This plot illustrates interaction coefficients when assessing binding of CR6261with H1H_1 (lower left triangle) and H9H_9 (upper right triangle).

    CR9114 @@ -280,7 +274,7 @@

    CR9114 theme(legend.position = "bottom") show(gg_CR9114)

    -

    This plot shows interaction coefficients when assessing binding of CR9114 with \(H_1\) (lower left triangle) and \(H_3\) (upper right triangle). Required mutations (indicated by R) are plotted in gray and left out of the analysis[^1].

    +

    This plot shows interaction coefficients when assessing binding of CR9114 with H1H_1 (lower left triangle) and H3H_3 (upper right triangle). Required mutations (indicated by R) are plotted in gray and left out of the analysis[^1].

    Our results show that mvMAPIT identifies all required mutations in these systems as well as most positions involved in at least one epistatic pair.

@@ -294,9 +288,7 @@

References - -

+ @@ -309,16 +301,16 @@

References

-

Site built with pkgdown 2.0.9.

+

Site built with pkgdown 2.1.1.

- - + + diff --git a/articles/study-wtccc-mice.html b/articles/study-wtccc-mice.html index 9858ab3..029ea77 100644 --- a/articles/study-wtccc-mice.html +++ b/articles/study-wtccc-mice.html @@ -6,12 +6,6 @@ Joint modeling of hematology traits yields epistatic signal in stock of mice • mvMAPIT - - - - - - @@ -28,7 +22,7 @@ - +
- +
@@ -114,7 +108,7 @@

Joint modeling of hematology traits yields epistatic signal in stock of mice

- Source: vignettes/study-wtccc-mice.Rmd + Source: vignettes/study-wtccc-mice.Rmd
@@ -131,12 +125,12 @@

Joint modeling of hematology traits yields epistatic signal in

Preprocessing of the heterogenous stock of mice dataset

-

This example study makes use of GWA data from the Wellcome Trust Centre for Human Genetics\(^{1,2,3}\) (http://mtweb.cs.ucl.ac.uk/mus/www/mouse/index.shtml). The genotypes from this study were downloaded directly using the BGLR-R package. This study contains \(N =\) 1,814 heterogenous stock of mice from 85 families (all descending from eight inbred progenitor strains)\(^{1,2}\), and 131 quantitative traits that are classified into 6 broad categories including behavior, diabetes, asthma, immunology, haematology, and biochemistry. Phenotypic measurements for these mice can be found freely available online to download (details can be found at http://mtweb.cs.ucl.ac.uk/mus/www/mouse/HS/index.shtml). In the main text, we focused on 15 hematological phenotypes including: atypical lymphocytes (ALY; Haem.ALYabs), basophils (BAS; Haem.BASabs), hematocrit (HCT; Haem.HCT), hemoglobin (HGB; Haem.HGB), large immature cells (LIC; Haem.LICabs), lymphocytes (LYM; Haem.LYMabs), mean corpuscular hemoglobin (MCH; Haem.MCH), mean corpuscular volume (MCV; Haem.MCV), monocytes (MON; Haem.MONabs), mean platelet volume (MPV; Haem.MPV), neutrophils (NEU; Haem.NEUabs), plateletcrit (PCT; Haem.PCT), platelets (PLT; Haem.PLT), red blood cell count (RBC; Haem.RBC), red cell distribution width (RDW; Haem.RDW), and white blood cell count (WBC; Haem.WBC). All phenotypes were previously corrected for sex, age, body weight, season, year, and cage effects \(^{1,2}\). For individuals with missing genotypes, we imputed values by the mean genotype of that SNP in their corresponding family. Only polymorphic SNPs with minor allele frequency above 5% were kept for the analyses. This left a total of \(J =\) 10,227 autosomal SNPs that were available for all mice.

+

This example study makes use of GWA data from the Wellcome Trust Centre for Human Genetics1,2,3^{1,2,3} (http://mtweb.cs.ucl.ac.uk/mus/www/mouse/index.shtml). The genotypes from this study were downloaded directly using the BGLR-R package. This study contains N=N = 1,814 heterogenous stock of mice from 85 families (all descending from eight inbred progenitor strains)1,2^{1,2}, and 131 quantitative traits that are classified into 6 broad categories including behavior, diabetes, asthma, immunology, haematology, and biochemistry. Phenotypic measurements for these mice can be found freely available online to download (details can be found at http://mtweb.cs.ucl.ac.uk/mus/www/mouse/HS/index.shtml). In the main text, we focused on 15 hematological phenotypes including: atypical lymphocytes (ALY; Haem.ALYabs), basophils (BAS; Haem.BASabs), hematocrit (HCT; Haem.HCT), hemoglobin (HGB; Haem.HGB), large immature cells (LIC; Haem.LICabs), lymphocytes (LYM; Haem.LYMabs), mean corpuscular hemoglobin (MCH; Haem.MCH), mean corpuscular volume (MCV; Haem.MCV), monocytes (MON; Haem.MONabs), mean platelet volume (MPV; Haem.MPV), neutrophils (NEU; Haem.NEUabs), plateletcrit (PCT; Haem.PCT), platelets (PLT; Haem.PLT), red blood cell count (RBC; Haem.RBC), red cell distribution width (RDW; Haem.RDW), and white blood cell count (WBC; Haem.WBC). All phenotypes were previously corrected for sex, age, body weight, season, year, and cage effects 1,2^{1,2}. For individuals with missing genotypes, we imputed values by the mean genotype of that SNP in their corresponding family. Only polymorphic SNPs with minor allele frequency above 5% were kept for the analyses. This left a total of J=J = 10,227 autosomal SNPs that were available for all mice.

Analyze hematology traits in mice

-

In this section, we apply mvMAPIT to individual-level genotypes and 15 hematology traits in a heterogeneous stock of mice dataset from the Wellcome Trust Centre for Human Genetics\(^{1,2,3}\). This collection of data contains approximately \(N =\) 2,000 individuals depending on the phenotype, and each mouse has been genotyped at \(J =\) 10,346 SNPs. Specifically, this stock of mice are known to be genetically related with population structure and the genetic architectures of these particular traits have been shown to have different levels of broad-sense heritability with varying contributions from non-additive genetic effects.

+

In this section, we apply mvMAPIT to individual-level genotypes and 15 hematology traits in a heterogeneous stock of mice dataset from the Wellcome Trust Centre for Human Genetics1,2,3^{1,2,3}. This collection of data contains approximately N=N = 2,000 individuals depending on the phenotype, and each mouse has been genotyped at J=J = 10,346 SNPs. Specifically, this stock of mice are known to be genetically related with population structure and the genetic architectures of these particular traits have been shown to have different levels of broad-sense heritability with varying contributions from non-additive genetic effects.

Apply mvMAPIT

@@ -147,12 +141,12 @@

Apply mvMAPITt(TRAIT$phenotype), test = "hybrid" )

-

As a result, we get redundant \(P\)-values for some of the univariate variance components. The statistical detection of epistasis is sensitive to sample size. Therefore, we coalesce the redundant data by keeping the analysis results of the largest data set used in the analysis and impute missing data from the next smaller data set that has no missing data.

+

As a result, we get redundant PP-values for some of the univariate variance components. The statistical detection of epistasis is sensitive to sample size. Therefore, we coalesce the redundant data by keeping the analysis results of the largest data set used in the analysis and impute missing data from the next smaller data set that has no missing data.

Analysis Data Availability

-

The results of the paper data are published on Harvard Dataverse. Find the files for Download here\(^{23}\). For running the code snippets in this vignette, download the two files

+

The results of the paper data are published on Harvard Dataverse. Find the files for Download here23^{23}. For running the code snippets in this vignette, download the two files

  • mice_HCTHGB_MCVMCH.rds
  • mice_SI_paper.rds
    • @@ -165,7 +159,7 @@

    Analysis Data Availability

    All Traits Overview

    -

    We also include results corresponding to the univariate MAPIT model and the covariance test for comparison. Overall, the single-trait marginal epistatic test does only identifies significant variants for the large immature cells (LIC) after Bonferroni correction (\(P = 4.83\times 10^{-6}\)). A complete picture of this can be seen in the following figure, which depicts Manhattan plots of our genome-wide interaction study for all combinations of trait pairs. Here, we can see that most of the signal in the combined \(P\)-values from mvMAPIT likely stems from the covariance component portion of the model.

    +

    We also include results corresponding to the univariate MAPIT model and the covariance test for comparison. Overall, the single-trait marginal epistatic test does only identifies significant variants for the large immature cells (LIC) after Bonferroni correction (P=4.83×10−6P = 4.83\times 10^{-6}). A complete picture of this can be seen in the following figure, which depicts Manhattan plots of our genome-wide interaction study for all combinations of trait pairs. Here, we can see that most of the signal in the combined PP-values from mvMAPIT likely stems from the covariance component portion of the model.

     for_ticks_chr <- aggregate(position ~ chr, mice_data$fisher, function(x) c(first = min(x), last = max(x))) %>%
   mutate(tick = floor((position[,"first"] + position[,"last"]) / 2)) %>%
@@ -212,7 +206,7 @@ 
    All Traits Overview
 
    Two trait pairs HCT & HGB as well as MCV & MCH
 
    
-
    The hypothesis that most of the signal in the combined \(P\)-values from mvMAPIT likely stems from the covariance component portion of the model holds true for the joint pairwise analysis of hematocrit (HCT) and hemoglobin (HGB) and mean corpuscular hemoglobin (MCH) and mean corpuscular volume (MCV) (e.g., see the third and fourth rows of the following figure).
    
+
    The hypothesis that most of the signal in the combined PP-values from mvMAPIT likely stems from the covariance component portion of the model holds true for the joint pairwise analysis of hematocrit (HCT) and hemoglobin (HGB) and mean corpuscular hemoglobin (MCH) and mean corpuscular volume (MCV) (e.g., see the third and fourth rows of the following figure).
    
 
     gg <- mice_HCTHGB_MCVMCH$fisher %>% ggplot(aes(
       x = position,
@@ -249,12 +243,12 @@ 
    Two trait pairs HCT & HG
                          labels = for_ticks_chr$chr2) +
       scale_linetype_manual(name = "", values = c('dashed'))
 show(gg)
    
-
    One explanation for observing more signal in the covariance components over the univariate test could be derived from the traits having low heritability but high correlation between epistatic interaction effects. In our simulation studies (see publication) we showed that the sensitivity of the covariance statistic increased for these cases. Notably, the non-additive signal identified by the covariance test is not totally dependent on the empirical correlation between traits. Instead, as previously shown in our simulation study, the power of mvMAPIT over the univariate approach occurs when there is correlation between the effects of epistatic interactions shared between two traits. Importantly, many of the candidate SNPs selected by the mvMAPIT framework have been previously discovered by past publications as having some functional nonlinear relationship with the traits of interest. For example, the multivariate analysis with traits MCH and MCV show a significant SNP rs4173870 (\(P = 4.89\times 10^{-10}\)) in the gene hematopoietic cell-specific Lyn substrate 1 (Hcls1) on chromosome 16 which has been shown to play a role in differentiation of erythrocytes\(^7\). Similarly, the joint analysis of HGB and HCT shows hits in multiple coding regions. One example here are the SNPs rs3692165 (\(P = 1.82\times 10^{-6}\)) and rs13482117 (\(P = 8.94\times 10^{-7}\)) in the gene calcium voltage-gated channel auxiliary subunit alpha2delta 3 (Cacna2d3) on chromosome 14, which has been associated with decreased circulating glucose levels\(^8\), and SNP rs3724260 (\(P = 4.58\times 10^{-6}\)) in the gene Dicer1 on chromosome 12 which has been annotated for anemia both in humans and mice\(^9\).
    
+
    One explanation for observing more signal in the covariance components over the univariate test could be derived from the traits having low heritability but high correlation between epistatic interaction effects. In our simulation studies (see publication) we showed that the sensitivity of the covariance statistic increased for these cases. Notably, the non-additive signal identified by the covariance test is not totally dependent on the empirical correlation between traits. Instead, as previously shown in our simulation study, the power of mvMAPIT over the univariate approach occurs when there is correlation between the effects of epistatic interactions shared between two traits. Importantly, many of the candidate SNPs selected by the mvMAPIT framework have been previously discovered by past publications as having some functional nonlinear relationship with the traits of interest. For example, the multivariate analysis with traits MCH and MCV show a significant SNP rs4173870 (P=4.89×10−10P = 4.89\times 10^{-10}) in the gene hematopoietic cell-specific Lyn substrate 1 (Hcls1) on chromosome 16 which has been shown to play a role in differentiation of erythrocytes7^7. Similarly, the joint analysis of HGB and HCT shows hits in multiple coding regions. One example here are the SNPs rs3692165 (P=1.82×10−6P = 1.82\times 10^{-6}) and rs13482117 (P=8.94×10−7P = 8.94\times 10^{-7}) in the gene calcium voltage-gated channel auxiliary subunit alpha2delta 3 (Cacna2d3) on chromosome 14, which has been associated with decreased circulating glucose levels8^8, and SNP rs3724260 (P=4.58×10−6P = 4.58\times 10^{-6}) in the gene Dicer1 on chromosome 12 which has been annotated for anemia both in humans and mice9^9.

    Notable SNPs with marginal epistatic effects after applying the mvMAPIT framework to 15 hematology traits

    -

    For full analysis, we provide a summary table which lists the combined \(P\)-values after running mvMAPIT with Fisher’s method. The following table lists a select subset of SNPs in coding regions of genes that have been associated with phenotypes related to the hematopoietic system, immune system, or homeostasis and metabolism. Each of these are significant (after correction for multiple hypothesis testing) in the mvMAPIT analysis of related hematology traits. Some of these phenotypes have been reported as having large broad-sense heritability, which improves the ability of mvMAPIT to detect the signal. For example, the genes Arf2 and Cacna2d3 are associated with phenotypes related to glucose homeostasis, which has been reported to have a large heritable component (estimated \(H^2 = 0.3\) for insulin sensitivity\(^{10}\)). Similarly, the genes App and Pex1 are associated with thrombosis where (an estimated) more than half of phenotypic variation has been attributed to genetic effects (estimated \(H^2 \ge 0.6\) for susceptibility to common thrombosis\(^{11}\)).

    +

    For full analysis, we provide a summary table which lists the combined PP-values after running mvMAPIT with Fisher’s method. The following table lists a select subset of SNPs in coding regions of genes that have been associated with phenotypes related to the hematopoietic system, immune system, or homeostasis and metabolism. Each of these are significant (after correction for multiple hypothesis testing) in the mvMAPIT analysis of related hematology traits. Some of these phenotypes have been reported as having large broad-sense heritability, which improves the ability of mvMAPIT to detect the signal. For example, the genes Arf2 and Cacna2d3 are associated with phenotypes related to glucose homeostasis, which has been reported to have a large heritable component (estimated H2=0.3H^2 = 0.3 for insulin sensitivity10^{10}). Similarly, the genes App and Pex1 are associated with thrombosis where (an estimated) more than half of phenotypic variation has been attributed to genetic effects (estimated H2≥0.6H^2 \ge 0.6 for susceptibility to common thrombosis11^{11}).

    @@ -274,10 +268,10 @@

    \(P\) -

    - - + + + + @@ -294,7 +288,7 @@

    \(^{12}\) +

    @@ -307,7 +301,7 @@

    \(^{13}\) +

    @@ -320,7 +314,7 @@

    \(^{14}\) +

    @@ -333,7 +327,7 @@

    \(^{14}\) +

    @@ -346,7 +340,7 @@

    \(^{15,16}\) +

    @@ -359,7 +353,7 @@

    \(^{15,16}\) +

    @@ -373,7 +367,7 @@

    \(^{17}\), \(^{18}\) +17^{17}, 18^{18}

    @@ -387,7 +381,7 @@

    \(^8\) +

    @@ -400,7 +394,7 @@

    \(^8\) +

    @@ -413,7 +407,7 @@

    \(^9\) +

    @@ -426,7 +420,7 @@

    \(^8\) +

    @@ -439,7 +433,7 @@

    \(^8\) +

    @@ -452,7 +446,7 @@

    \(^8\) +

    @@ -465,7 +459,7 @@

    \(^8\) +

    @@ -478,7 +472,7 @@

    \(^8\) +

    @@ -491,7 +485,7 @@

    \(^{19}\) +

    @@ -504,7 +498,7 @@

    \(^{19}\) +

    @@ -517,7 +511,7 @@

    \(^7\) +

    @@ -530,7 +524,7 @@

    \(^{20,11}\) +

    @@ -543,7 +537,7 @@

    \(^{20,11}\) +

    @@ -556,11 +550,11 @@

    \(^{21}\) +

    Trait 2 \(P\)Cov. \(P\)Comb. \(P\)Trait 1 PPTrait 2 PPCov. PPComb. PP Gene Genomic Annotation Reference12^{12}
    rs1347809213^{13}
    rs369488714^{14}
    rs369488714^{14}
    rs1347892315,16^{15,16}
    rs1347892415,16^{15,16}
    rs13478985
    8^8
    rs37231638^8
    rs37242609^9
    rs36921658^8
    rs36974668^8
    rs134821178^8
    rs61597868^8
    rs62445698^8
    rs1348228819^{19}
    rs368044819^{19}
    rs41738707^7
    rs421210220,11^{20,11}
    rs421218620,11^{20,11}
    rs371199421^{21}
    -

    In the first two columns, we list SNPs and their genetic location according to the mouse assembly NCBI build 34 (accessed from \(^{21}\)) in the format Chromosome:Basepair. Next, we give the results stemming from univariate analyses on traits 1 and 2, respectively, the covariance (cov) test, and the overall \(P\)-value derived by mvMAPIT using Fisher’s method. The last columns detail the closest neighboring genes found using the Mouse Genome Informatics database\(^4\) \(^5\) \(^6\), a short summary of the suggested annotated function for those genes, and the reference to the source of the annotation.

    +

    In the first two columns, we list SNPs and their genetic location according to the mouse assembly NCBI build 34 (accessed from 21^{21}) in the format Chromosome:Basepair. Next, we give the results stemming from univariate analyses on traits 1 and 2, respectively, the covariance (cov) test, and the overall PP-value derived by mvMAPIT using Fisher’s method. The last columns detail the closest neighboring genes found using the Mouse Genome Informatics database4^45^56^6, a short summary of the suggested annotated function for those genes, and the reference to the source of the annotation.

@@ -594,9 +588,7 @@

References - -

+

@@ -609,16 +601,16 @@

References

-

Site built with pkgdown 2.0.9.

+

Site built with pkgdown 2.1.1.

- - + + diff --git a/articles/tutorial-docker-mvmapit.html b/articles/tutorial-docker-mvmapit.html index 6914b6f..a5a2301 100644 --- a/articles/tutorial-docker-mvmapit.html +++ b/articles/tutorial-docker-mvmapit.html @@ -6,12 +6,6 @@ Dockerized mvMAPIT • mvMAPIT - - - - - - @@ -28,7 +22,7 @@ - +
- +
@@ -114,7 +108,7 @@

Dockerized mvMAPIT

- Source: vignettes/tutorial-docker-mvmapit.Rmd + Source: vignettes/tutorial-docker-mvmapit.Rmd
@@ -150,14 +144,14 @@

Run the mvMAPIT Imagemvmapit(t(simulated_data$genotype[1:100,1:10]), t(simulated_data$trait[1:100,]), cores = 2, logLevel = "DEBUG")

-
## 2024-06-07 08:32:21.344939 DEBUG:mvmapit:Running in normal test mode.
-## 2024-06-07 08:32:21.35245 DEBUG:mvmapit:Genotype matrix: 10 x 100
-## 2024-06-07 08:32:21.363777 DEBUG:mvmapit:Phenotype matrix: 2 x 100
-## 2024-06-07 08:32:21.364519 DEBUG:mvmapit:Number of zero variance variants: 0
-## 2024-06-07 08:32:21.365189 DEBUG:mvmapit:Genotype matrix after removing zero variance variants: 10 x 100
-## 2024-06-07 08:32:21.365666 DEBUG:mvmapit:Scale X matrix appropriately.
-## 2024-06-07 08:32:21.366654 INFO:mvmapit:Running normal C++ routine.
-## 2024-06-07 08:32:21.381684 DEBUG:mvmapit:Calculated mean time of execution. Return list.
+
## 2024-10-21 17:58:15.211364 DEBUG:mvmapit:Running in normal test mode.
+## 2024-10-21 17:58:15.218951 DEBUG:mvmapit:Genotype matrix: 10 x 100
+## 2024-10-21 17:58:15.230692 DEBUG:mvmapit:Phenotype matrix: 2 x 100
+## 2024-10-21 17:58:15.23145 DEBUG:mvmapit:Number of zero variance variants: 0
+## 2024-10-21 17:58:15.232141 DEBUG:mvmapit:Genotype matrix after removing zero variance variants: 10 x 100
+## 2024-10-21 17:58:15.232651 DEBUG:mvmapit:Scale X matrix appropriately.
+## 2024-10-21 17:58:15.23364 INFO:mvmapit:Running normal C++ routine.
+## 2024-10-21 17:58:15.254921 DEBUG:mvmapit:Calculated mean time of execution. Return list.
## $pvalues
 ## # A tibble: 30 × 3
 ##    id        trait         p
@@ -193,9 +187,9 @@ 
Run the mvMAPIT Image## $duration
 ##       process duration_ms
 ## 1         cov           0
-## 2 projections           1
+## 2 projections           3
 ## 3   vectorize           0
-## 4           q           0
+## 4           q           3
 ## 5           S           0
 ## 6          vc           0
@@ -203,9 +197,7 @@

Run the mvMAPIT Image - - + @@ -218,16 +210,16 @@

Run the mvMAPIT Image

-

Site built with pkgdown 2.0.9.

+

Site built with pkgdown 2.1.1.

- - + + diff --git a/articles/tutorial-lt-mapit.html b/articles/tutorial-lt-mapit.html index 298876d..38cfd87 100644 --- a/articles/tutorial-lt-mapit.html +++ b/articles/tutorial-lt-mapit.html @@ -6,12 +6,6 @@ Liability threshold MAPIT • mvMAPIT - - - - - - @@ -28,7 +22,7 @@ - +
- +
@@ -114,7 +108,7 @@

Liability threshold MAPIT

- Source: vignettes/tutorial-lt-mapit.Rmd + Source: vignettes/tutorial-lt-mapit.Rmd
@@ -123,7 +117,7 @@

Liability threshold MAPIT

-

In this tutorial we will illustrate how to use the Liability Threshold MArginal ePIstasis Test (LT-MAPIT) by Crawford and Zhou (2018)\(^1\). For this purpose we will first simulate synthetic data and then analyze it.

+

In this tutorial we will illustrate how to use the Liability Threshold MArginal ePIstasis Test (LT-MAPIT) by Crawford and Zhou (2018)1^1. For this purpose we will first simulate synthetic data and then analyze it.

The data are single nucleotide polymorphisms (SNPs) with simulated genotypes. For the simulation we choose the following set of parameters:

  1. @@ -131,9 +125,9 @@

    Liability threshold MAPIT

  2. n_snps - number of SNPs or variants
  3. -PVE - phenotypic variance explained/broad-sense heritability (\(H^2\))
    4. +PVE - phenotypic variance explained/broad-sense heritability (H2H^2)
  4. -rho - measures the portion of \(H^2\) that is contributed by the marignal (additive) effects
    5. +rho - measures the portion of H2H^2 that is contributed by the marignal (additive) effects
  5. disease_prevalence - assumed disease prevelance in the population
  6. @@ -207,9 +201,7 @@

    References - -

+
@@ -222,16 +214,16 @@

References

-

Site built with pkgdown 2.0.9.

+

Site built with pkgdown 2.1.1.

- - + + diff --git a/articles/tutorial-simulations.html b/articles/tutorial-simulations.html index 830fe33..5d5881c 100644 --- a/articles/tutorial-simulations.html +++ b/articles/tutorial-simulations.html @@ -6,12 +6,6 @@ Simulate Traits • mvMAPIT - - - - - - @@ -28,7 +22,7 @@ - +
- +
@@ -114,7 +108,7 @@

Simulate Traits

- Source: vignettes/tutorial-simulations.Rmd + Source: vignettes/tutorial-simulations.Rmd
@@ -123,7 +117,7 @@

Simulate Traits

-

Data are simulated sample genotypes with \(p\) single nucleotide polymorphisms (SNPs) and \(n\) samples togeter with some quantitative traits.

+

Data are simulated sample genotypes with pp single nucleotide polymorphisms (SNPs) and nn samples togeter with some quantitative traits.

Simulate Genotypes

@@ -150,11 +144,11 @@

Simulate Genotypes\(0\) if the genotype is homozygous in the reference allele, +00 if the genotype is homozygous in the reference allele,
  • -\(1\) if both, the reference allele and the alternative allele are present,
    • +11 if both, the reference allele and the alternative allele are present,
  • -\(2\) if the genotype is homozygous in the alternative allele.
    • +22 if the genotype is homozygous in the alternative allele.

    An example genotype matrix could look like this:

    @@ -210,9 +204,9 @@

    Simulate Traits d: number of traits
  • -PVE: phenotypic variance explained/broad-sense heritability (\(H^2\))
    • +PVE: phenotypic variance explained/broad-sense heritability (H2H^2)
  • -rho: portion of \(H^2\) that is contributed by the marginal (additive) effects
    • +rho: portion of H2H^2 that is contributed by the marginal (additive) effects
  • Number of SNPs that are causing variation in the trait:
    • @@ -270,36 +264,34 @@

      Simulate TraitsDetails on Simulating Pairwise Epistasis

      The general simulation scheme for interactions follows the methods outlined in Crawford et al. (2017). For the data representing the null hypothesis of no epistasis present, 1000 SNPs without rare variants are sampled from the genotype data. The effect sizes of the causal SNPs then are sampled independently from a multivariate normal distribution according to equation .

      -

      \[\begin{equation}\label{eq:null_simulations_beta} +

      𝐛i∼𝒩d(0,𝐔)\begin{equation}\label{eq:null_simulations_beta} \mathbf{b}_i \sim \mathcal{N}_{d}\left(0, \mathbf{U}\right) -\end{equation}\]

      -

      As parameter of the simulations, \(\mathbf{U}\) describes the covariance between the additive effects of the causal SNPs on the \(d\) different traits. The effects of all other SNPs are set to zero.

      +\end{equation}

      +

      As parameter of the simulations, 𝐔\mathbf{U} describes the covariance between the additive effects of the causal SNPs on the dd different traits. The effects of all other SNPs are set to zero.

      In order to simulate pleiotropy, the number of desired pleiotropic SNPs is sampled from the set of causal SNPs. These SNPs are then included in simulating interactions for every trait. Additionally, trait specific SNPs are sampled independently for each trait from the causal SNPs without the pleiotropic SNPs.

      -

      The interactions are modeled by selecting the number of desired SNPs that have non-zero pairwise interaction effects. These SNPs then are split into two groups. Each SNP from one group is simulated to interact with each SNP of the other group but not with any SNP within the same group that it was assigned to. This results in \(n_1 \cdot n_2\) epistatic interactions, with \(n_i\) the number of SNPs in group i. For all these interactions, random effects are sampled from the multivariate normal distribution given in .

      -

      \[\begin{equation}\label{eq:simulations_alpha} +

      The interactions are modeled by selecting the number of desired SNPs that have non-zero pairwise interaction effects. These SNPs then are split into two groups. Each SNP from one group is simulated to interact with each SNP of the other group but not with any SNP within the same group that it was assigned to. This results in n1⋅n2n_1 \cdot n_2 epistatic interactions, with nin_i the number of SNPs in group i. For all these interactions, random effects are sampled from the multivariate normal distribution given in .

      +

      𝐚i∼𝒩d(0,𝐕)\begin{equation}\label{eq:simulations_alpha} \mathbf{a}_i \sim \mathcal{N}_{d}\left(0, \mathbf{V}\right) -\end{equation}\]

      -

      Analogously, \(\mathbf{V}\) is the covariance between the pariwise epistatic effects of the causal SNPs on the \(d\) different traits. The portion of the variance \(\mathbf{E}\) that is not due to heritable effects is simulated as multivariate normal with no correlation. The effects of the SNPs and interactions are collected in matrices \(\mathbf{A}\) and \(\mathbf{B}\). The matrices \(\mathbf{X}\) and \(\mathbf{W}\) contain the genotype data of the causal SNPs and interactions. With these effects, the phenotypes then are simulated according to equation .

      -

      \[\begin{equation}\label{eq:simulation_mapit} +\end{equation}

      +

      Analogously, 𝐕\mathbf{V} is the covariance between the pariwise epistatic effects of the causal SNPs on the dd different traits. The portion of the variance 𝐄\mathbf{E} that is not due to heritable effects is simulated as multivariate normal with no correlation. The effects of the SNPs and interactions are collected in matrices 𝐀\mathbf{A} and 𝐁\mathbf{B}. The matrices 𝐗\mathbf{X} and 𝐖\mathbf{W} contain the genotype data of the causal SNPs and interactions. With these effects, the phenotypes then are simulated according to equation .

      +

      𝐘=𝐗𝐁+𝐖𝐀+𝐄\begin{equation}\label{eq:simulation_mapit} \mathbf{Y} = \mathbf{X}\mathbf{B} + \mathbf{W}\mathbf{A} + \mathbf{E} -\end{equation}\]

      +\end{equation}

      For each trait independently, the variance components are scaled such that the variance can be partitioned according to equation .

      -

      \[\begin{align}\label{eq:simulation_var_comp} +

      𝐘=𝐘X+𝐘W+𝐄var(𝐘)=var(𝐘X)+var(𝐘W)+var(𝐄)=ρ⋅H2+(1−ρ)⋅H2+(1−H2)=1\begin{align}\label{eq:simulation_var_comp} \mathbf{Y} &= \mathbf{Y}_{X} + \mathbf{Y}_{W} + \mathbf{E} \\ \mathrm{var}(\mathbf{Y}) &= \mathrm{var}(\mathbf{Y}_{X}) + \mathrm{var}(\mathbf{Y}_{W}) + \mathrm{var}(\mathbf{E}) \\ &= \rho \cdot H^2 + (1 - \rho) \cdot H^2 + (1 - H^2) \\ &= 1 -\end{align}\]

      +\end{align}

      + @@ -312,16 +304,16 @@

      Details on Simulating Pairwise

      -

      Site built with pkgdown 2.0.9.

      +

      Site built with pkgdown 2.1.1.

      - - + + diff --git a/authors.html b/authors.html index 01192e1..2c41874 100644 --- a/authors.html +++ b/authors.html @@ -1,9 +1,9 @@ -Authors and Citation • mvMAPITAuthors and Citation • mvMAPIT - +
      - +
      @@ -79,7 +79,7 @@

      Authors and Citation

      - +

      • Julian Stamp. Maintainer, author.

        @@ -92,14 +92,14 @@

        Authors and Citation

        Citation

        - Source: DESCRIPTION + Source: DESCRIPTION

        Stamp J, Crawford L (2024). mvMAPIT: Multivariate Genome Wide Marginal Epistasis Test. -R package version 2.0.3, https://lcrawlab.github.io/mvMAPIT/, https://github.com/lcrawlab/mvMAPIT. +R package version 2.0.3, https://lcrawlab.github.io/mvMAPIT/, https://github.com/lcrawlab/mvMAPIT. 

        @Manual{,
   title = {mvMAPIT: Multivariate Genome Wide Marginal Epistasis Test},
@@ -120,15 +120,15 @@ 
        Citation
      -

      Site built with pkgdown 2.0.9.

      +

      Site built with pkgdown 2.1.1.

      - - + + diff --git a/favicon-16x16.png b/favicon-16x16.png deleted file mode 100644 index 3bacc38..0000000 Binary files a/favicon-16x16.png and /dev/null differ diff --git a/favicon-32x32.png b/favicon-32x32.png deleted file mode 100644 index ccfd6f9..0000000 Binary files a/favicon-32x32.png and /dev/null differ diff --git a/favicon.ico b/favicon.ico deleted file mode 100644 index eb409e0..0000000 Binary files a/favicon.ico and /dev/null differ diff --git a/index.html b/index.html index 4692026..e2282f6 100644 --- a/index.html +++ b/index.html @@ -6,12 +6,6 @@ Multivariate Genome Wide Marginal Epistasis Test • mvMAPIT - - - - - - @@ -27,7 +21,7 @@ - +
      - +
      @@ -113,7 +107,7 @@

      Multivariate MAPIT Documentation

      -

      R CMD check Docker Image CI CRAN downloads CRAN_Status_Badge

      +

      R CMD check Docker Image CI CRAN downloads CRAN_Status_Badge

      Find the full package documentation including examples and articles here: Multivariate MAPIT Documentation.

      The multivariate MArginal ePIstasis Test (mvMAPIT) @@ -277,16 +271,16 @@

      Developers

      -

      Site built with pkgdown 2.0.9.

      +

      Site built with pkgdown 2.1.1.

      - - + + diff --git a/news/index.html b/news/index.html index d1314f5..7380938 100644 --- a/news/index.html +++ b/news/index.html @@ -1,9 +1,9 @@ -Changelog • mvMAPITChangelog • mvMAPIT - +
      - +

      Changelog

      - Source: NEWS.md + Source: NEWS.md
      @@ -131,15 +131,15 @@

      -

      Site built with pkgdown 2.0.9.

      +

      Site built with pkgdown 2.1.1.

      - - + + diff --git a/pkgdown.yml b/pkgdown.yml index 4974479..0d5473a 100644 --- a/pkgdown.yml +++ b/pkgdown.yml @@ -1,5 +1,5 @@ pandoc: 2.9.2.1 -pkgdown: 2.0.9 +pkgdown: 2.1.1 pkgdown_sha: ~ articles: mvMAPIT: mvMAPIT.html @@ -9,8 +9,7 @@ articles: tutorial-docker-mvmapit: tutorial-docker-mvmapit.html tutorial-lt-mapit: tutorial-lt-mapit.html tutorial-simulations: tutorial-simulations.html -last_built: 2024-06-07T08:30Z +last_built: 2024-10-21T17:56Z urls: reference: https://lcrawlab.github.io/mvMAPIT/reference article: https://lcrawlab.github.io/mvMAPIT/articles - diff --git a/reference/Rplot001.png b/reference/Rplot001.png deleted file mode 100644 index 17a3580..0000000 Binary files a/reference/Rplot001.png and /dev/null differ diff --git a/reference/binary_to_liability.html b/reference/binary_to_liability.html index 0c9d874..60e1157 100644 --- a/reference/binary_to_liability.html +++ b/reference/binary_to_liability.html @@ -1,13 +1,13 @@ -Convert binary traits to liabilities for low prevalence — binary_to_liability • mvMAPITConvert binary traits to liabilities for low prevalence — binary_to_liability • mvMAPIT - +
      - +

      Convert binary traits to liabilities for low prevalence

      - Source: R/binary_to_liability.R + Source: R/binary_to_liability.R

      This function implements an approximate conversion of binary traits to liabilties as proposed in the LT-MAPIT model (Crawford and Zhou 2018, -https://doi.org/10.1101/374983). Note that this is only good for low prevalence. +https://doi.org/10.1101/374983). Note that this is only good for low prevalence. To run LT-MAPIT (MAPIT on case-control traits), convert the binary traits to liabilities using this function and pass the liabilities to mvmapit as trait.

      @@ -98,19 +98,19 @@

      Convert binary traits to liabilities for low prevalence

      Arguments

      -
      case_control_trait
      + + +
      case_control_trait

      Case-control trait encoded as binary trait with 0 as control or 1 as case.

      -
      prevalence
      +
      prevalence

      Case prevalence between 0 and 1. Proportion of cases in the population.

      Value

      - - -

      A trait vector of same length as y with case-control indicators converted +

      A trait vector of same length as y with case-control indicators converted to liabilties.

      @@ -126,15 +126,15 @@

      Value

      -

      Site built with pkgdown 2.0.9.

      +

      Site built with pkgdown 2.1.1.

      - - + + diff --git a/reference/cauchy_combined.html b/reference/cauchy_combined.html index 4f6b360..52dd3dc 100644 --- a/reference/cauchy_combined.html +++ b/reference/cauchy_combined.html @@ -1,10 +1,10 @@ -Cauchy p combine method on mvmapit return — cauchy_combined • mvMAPITCauchy p combine method on mvmapit return — cauchy_combined • mvMAPIT - +
      - +

      Cauchy p combine method on mvmapit return

      - Source: R/combine_pvalues.R + Source: R/combine_pvalues.R
      @@ -92,25 +92,25 @@

      Cauchy p combine method on mvmapit return

      Arguments

      -
      pvalues
      + + +
      pvalues

      Tibble with p-values from mvmapit function call. Grouping is based on the column named "id"

      -
      group_col
      +
      group_col

      String that denotes column by which to group and combine p-values.

      -
      p_col
      +
      p_col

      String that denotes p-value column.

      Value

      - - -

      A Tibble with the combined p-values.

      +

      A Tibble with the combined p-values.

      @@ -132,7 +132,7 @@

      Examples

      t(Y), test = "normal", cores = 1, logLevel = "INFO" ) -#> 2024-06-07 08:30:55.234871 INFO:mvmapit:Running normal C++ routine. +#> 2024-10-21 17:56:44.373648 INFO:mvmapit:Running normal C++ routine. cauchy <- cauchy_combined(mapit$pvalues)
      @@ -148,15 +148,15 @@

      Examples

      -

      Site built with pkgdown 2.0.9.

      +

      Site built with pkgdown 2.1.1.

      - - + + diff --git a/reference/fishers_combined.html b/reference/fishers_combined.html index ca116c7..f0f6e83 100644 --- a/reference/fishers_combined.html +++ b/reference/fishers_combined.html @@ -1,10 +1,10 @@ -Fisher's combine method on mvmapit return — fishers_combined • mvMAPITFisher's combine method on mvmapit return — fishers_combined • mvMAPIT - +
      - +

      Fisher's combine method on mvmapit return

      - Source: R/combine_pvalues.R + Source: R/combine_pvalues.R
      @@ -92,24 +92,24 @@

      Fisher's combine method on mvmapit return

      Arguments

      -
      pvalues
      + + +
      pvalues

      Tibble with p-values from mvmapit function call.

      -
      group_col
      +
      group_col

      String that denotes column by which to group and combine p-values.

      -
      p_col
      +
      p_col

      String that denotes p-value column.

      Value

      - - -

      A Tibble with the combined p-values.

      +

      A Tibble with the combined p-values.

      @@ -131,7 +131,7 @@

      Examples

      t(Y), test = "normal", cores = 1, logLevel = "INFO" ) -#> 2024-06-07 08:30:55.694391 INFO:mvmapit:Running normal C++ routine. +#> 2024-10-21 17:56:45.042037 INFO:mvmapit:Running normal C++ routine. fisher <- fishers_combined(mapit$pvalues)
      @@ -147,15 +147,15 @@

      Examples

      -

      Site built with pkgdown 2.0.9.

      +

      Site built with pkgdown 2.1.1.

      - - + + diff --git a/reference/harmonic_combined.html b/reference/harmonic_combined.html index c2903a9..c92300d 100644 --- a/reference/harmonic_combined.html +++ b/reference/harmonic_combined.html @@ -1,10 +1,10 @@ -Harmonic mean p combine method on mvmapit return — harmonic_combined • mvMAPITHarmonic mean p combine method on mvmapit return — harmonic_combined • mvMAPIT - +
      - +

      Harmonic mean p combine method on mvmapit return

      - Source: R/combine_pvalues.R + Source: R/combine_pvalues.R
      @@ -92,25 +92,25 @@

      Harmonic mean p combine method on mvmapit return

      Arguments

      -
      pvalues
      + + +
      pvalues

      Tibble with p-values from mvmapit function call. Grouping is based on the column named "id"

      -
      group_col
      +
      group_col

      String that denotes column by which to group and combine p-values.

      -
      p_col
      +
      p_col

      String that denotes p-value column.

      Value

      - - -

      A Tibble with the combined p-values.

      +

      A Tibble with the combined p-values.

      @@ -132,7 +132,7 @@

      Examples

      t(Y), test = "normal", cores = 1, logLevel = "INFO" ) -#> 2024-06-07 08:30:56.035134 INFO:mvmapit:Running normal C++ routine. +#> 2024-10-21 17:56:45.565575 INFO:mvmapit:Running normal C++ routine. harmonic <- harmonic_combined(mapit$pvalues)
      @@ -148,15 +148,15 @@

      Examples

      -

      Site built with pkgdown 2.0.9.

      +

      Site built with pkgdown 2.1.1.

      - - + + diff --git a/reference/index.html b/reference/index.html index 2fa44c9..f47f221 100644 --- a/reference/index.html +++ b/reference/index.html @@ -1,9 +1,9 @@ -Function reference • mvMAPITPackage index • mvMAPIT - +
      - +
      @@ -151,15 +151,15 @@

      Real Data Examples
      -

      Site built with pkgdown 2.0.9.

      +

      Site built with pkgdown 2.1.1.

      - - + + diff --git a/reference/mvmapit.html b/reference/mvmapit.html index 638c227..277f2d3 100644 --- a/reference/mvmapit.html +++ b/reference/mvmapit.html @@ -1,10 +1,10 @@ -Multivariate MArginal ePIstasis Test (mvMAPIT) — mvmapit • mvMAPITMultivariate MArginal ePIstasis Test (mvMAPIT) — mvmapit • mvMAPIT - +
      - +

      Multivariate MArginal ePIstasis Test (mvMAPIT)

      - Source: R/MAPIT.R + Source: R/MAPIT.R
      @@ -104,55 +104,55 @@

      Multivariate MArginal ePIstasis Test (mvMAPIT)

      Arguments

      -
      X
      + + +
      X

      is the p x n genotype matrix where p is the number of variants and n is the number of samples. Must be a matrix and not a data.frame.

      -
      Y
      +
      Y

      is the d x n matrix of d quantitative or continuous traits for n samples.

      -
      Z
      +
      Z

      is the matrix q x n matrix of covariates. Must be a matrix and not a data.frame.

      -
      C
      +
      C

      is an n x n covariance matrix detailing environmental effects and population structure effects.

      -
      threshold
      +
      threshold

      is a parameter detailing the value at which to recalibrate the Z test p values. If nothing is defined by the user, the default value will be 0.05 as recommended by the Crawford et al. (2017).

      -
      accuracy
      +
      accuracy

      is a parameter setting the davies function numerical approximation accuracy. This parameter is not needed for the normal test. Smaller p-values than the accuracy will be zero.

      -
      test
      +
      test

      is a parameter defining what hypothesis test should be run. Takes on values 'normal', 'davies', and 'hybrid'. The 'hybrid' test runs first the 'normal' test and then the 'davies' test on the significant variants.

      -
      cores
      +
      cores

      is a parameter detailing the number of cores to parallelize over. It is important to note that this value only matters when the user has installed OPENMP on their operating system.

      -
      variantIndex
      +
      variantIndex

      is a vector containing indices of variants to be included in the computation.

      -
      logLevel
      +
      logLevel

      is a string parameter defining the log level for the logging package.

      -
      logFile
      +
      logFile

      is a string parameter defining the name of the log file for the logging output. Default is stdout.

      Value

      - - -

      A list of P values and PVEs

      +

      A list of P values and PVEs

      Details

      @@ -202,7 +202,7 @@

      Examples

      t(Y), test = "normal", cores = 1, logLevel = "INFO" ) -#> 2024-06-07 08:30:56.37413 INFO:mvmapit:Running normal C++ routine. +#> 2024-10-21 17:56:46.100952 INFO:mvmapit:Running normal C++ routine.
      @@ -217,15 +217,15 @@

      Examples

      -

      Site built with pkgdown 2.0.9.

      +

      Site built with pkgdown 2.1.1.

      - - + + diff --git a/reference/mvmapit_data.html b/reference/mvmapit_data.html index 3db885b..d585ee2 100644 --- a/reference/mvmapit_data.html +++ b/reference/mvmapit_data.html @@ -1,11 +1,11 @@ -Multivariate MAPIT analysis and exhaustive search analysis. — mvmapit_data • mvMAPITMultivariate MAPIT analysis and exhaustive search analysis. — mvmapit_data • mvMAPIT - +
      - +

      Multivariate MAPIT analysis and exhaustive search analysis.

      - Source: R/mvmapit_data.R + Source: R/mvmapit_data.R
      @@ -122,15 +122,15 @@

      Source

      -

      Site built with pkgdown 2.0.9.

      +

      Site built with pkgdown 2.1.1.

      - - + + diff --git a/reference/phillips_data.html b/reference/phillips_data.html index d9387d9..b57c096 100644 --- a/reference/phillips_data.html +++ b/reference/phillips_data.html @@ -1,11 +1,11 @@ -Multivariate MAPIT analysis of binding affinities in broadly neutralizing antibodies. — phillips_data • mvMAPITMultivariate MAPIT analysis of binding affinities in broadly neutralizing antibodies. — phillips_data • mvMAPIT - +
      - +

      Multivariate MAPIT analysis of binding affinities in broadly neutralizing antibodies.

      - Source: R/phillips_data.R + Source: R/phillips_data.R
      @@ -130,15 +130,15 @@

      Details

      -

      Site built with pkgdown 2.0.9.

      +

      Site built with pkgdown 2.1.1.

      - - + + diff --git a/reference/simulate_traits.html b/reference/simulate_traits.html index a05da1b..7d82f49 100644 --- a/reference/simulate_traits.html +++ b/reference/simulate_traits.html @@ -1,9 +1,9 @@ -Simulate phenotye data — simulate_traits • mvMAPITSimulate phenotye data — simulate_traits • mvMAPIT - +
      - +

      Simulate phenotye data

      - Source: R/simulate_data.R + Source: R/simulate_data.R
      @@ -106,71 +106,71 @@

      Simulate phenotye data

      Arguments

      -
      genotype_matrix
      + + +
      genotype_matrix

      Genotype matrix with samples as rows, and SNPs as columns.

      -
      n_causal
      +
      n_causal

      Number of SNPs that are causal.

      -
      n_trait_specific
      +
      n_trait_specific

      Number of causal SNPs with single trait epistatic effects.

      -
      n_pleiotropic
      +
      n_pleiotropic

      Number of SNPs with epistatic effects on all traits.

      -
      H2
      +
      H2

      Broad-sense heritability. Can be vector.

      -
      d
      +
      d

      Number of traits.

      -
      rho
      +
      rho

      Proportion of heritability explained by additivity.

      -
      marginal_correlation
      +
      marginal_correlation

      Correlation between the additive effects of the trait.

      -
      epistatic_correlation
      +
      epistatic_correlation

      Correlation between the epistatic effects of the trait.

      -
      group_ratio_trait
      +
      group_ratio_trait

      Ratio of sizes of trait specific groups that interact, e.g. a ratio 1:3 would be value 3.

      -
      group_ratio_pleiotropic
      +
      group_ratio_pleiotropic

      Ratio of sizes of pleiotropic groups that interact, e.g. a ratio 1:3 would be value 3.

      -
      maf_threshold
      +
      maf_threshold

      is a float parameter defining the threshold for the minor allele frequency not included in causal SNPs.

      -
      seed
      +
      seed

      Random seed for simulation.

      -
      logLevel
      +
      logLevel

      is a string parameter defining the log level for the logging package.

      -
      logFile
      +
      logFile

      is a string parameter defining the name of the log file for the logging output.

      Value

      - - -

      A list object containing the trait data, the genotype data, as well as the causal SNPs and summary statistics.

      +

      A list object containing the trait data, the genotype data, as well as the causal SNPs and summary statistics.

      Details

      @@ -208,15 +208,15 @@

      Examples

      -

      Site built with pkgdown 2.0.9.

      +

      Site built with pkgdown 2.1.1.

      - - + + diff --git a/reference/simulated_data.html b/reference/simulated_data.html index 0d4d721..f41940b 100644 --- a/reference/simulated_data.html +++ b/reference/simulated_data.html @@ -1,9 +1,9 @@ -Genotype and trait data with epistasis. — simulated_data • mvMAPITGenotype and trait data with epistasis. — simulated_data • mvMAPIT - +
      - +

      Genotype and trait data with epistasis.

      - Source: R/simulated_data.R + Source: R/simulated_data.R
      @@ -130,15 +130,15 @@

      Source

      -

      Site built with pkgdown 2.0.9.

      +

      Site built with pkgdown 2.1.1.

      - - + + diff --git a/sitemap.xml b/sitemap.xml index 9a07b11..626ce9a 100644 --- a/sitemap.xml +++ b/sitemap.xml @@ -1,72 +1,26 @@ - - - - https://lcrawlab.github.io/mvMAPIT/404.html - - - https://lcrawlab.github.io/mvMAPIT/LICENSE.html - - - https://lcrawlab.github.io/mvMAPIT/articles/index.html - - - https://lcrawlab.github.io/mvMAPIT/articles/mvMAPIT.html - - - https://lcrawlab.github.io/mvMAPIT/articles/study-compare-p-value-combine-methods.html - - - https://lcrawlab.github.io/mvMAPIT/articles/study-phillips-bnabs.html - - - https://lcrawlab.github.io/mvMAPIT/articles/study-wtccc-mice.html - - - https://lcrawlab.github.io/mvMAPIT/articles/tutorial-docker-mvmapit.html - - - https://lcrawlab.github.io/mvMAPIT/articles/tutorial-lt-mapit.html - - - https://lcrawlab.github.io/mvMAPIT/articles/tutorial-simulations.html - - - https://lcrawlab.github.io/mvMAPIT/authors.html - - - https://lcrawlab.github.io/mvMAPIT/index.html - - - https://lcrawlab.github.io/mvMAPIT/news/index.html - - - https://lcrawlab.github.io/mvMAPIT/reference/binary_to_liability.html - - - https://lcrawlab.github.io/mvMAPIT/reference/cauchy_combined.html - - - https://lcrawlab.github.io/mvMAPIT/reference/fishers_combined.html - - - https://lcrawlab.github.io/mvMAPIT/reference/harmonic_combined.html - - - https://lcrawlab.github.io/mvMAPIT/reference/index.html - - - https://lcrawlab.github.io/mvMAPIT/reference/mvmapit.html - - - https://lcrawlab.github.io/mvMAPIT/reference/mvmapit_data.html - - - https://lcrawlab.github.io/mvMAPIT/reference/phillips_data.html - - - https://lcrawlab.github.io/mvMAPIT/reference/simulate_traits.html - - - https://lcrawlab.github.io/mvMAPIT/reference/simulated_data.html - + +https://lcrawlab.github.io/mvMAPIT/404.html +https://lcrawlab.github.io/mvMAPIT/LICENSE.html +https://lcrawlab.github.io/mvMAPIT/articles/index.html +https://lcrawlab.github.io/mvMAPIT/articles/mvMAPIT.html +https://lcrawlab.github.io/mvMAPIT/articles/study-compare-p-value-combine-methods.html +https://lcrawlab.github.io/mvMAPIT/articles/study-phillips-bnabs.html +https://lcrawlab.github.io/mvMAPIT/articles/study-wtccc-mice.html +https://lcrawlab.github.io/mvMAPIT/articles/tutorial-docker-mvmapit.html +https://lcrawlab.github.io/mvMAPIT/articles/tutorial-lt-mapit.html +https://lcrawlab.github.io/mvMAPIT/articles/tutorial-simulations.html +https://lcrawlab.github.io/mvMAPIT/authors.html +https://lcrawlab.github.io/mvMAPIT/index.html +https://lcrawlab.github.io/mvMAPIT/news/index.html +https://lcrawlab.github.io/mvMAPIT/reference/binary_to_liability.html +https://lcrawlab.github.io/mvMAPIT/reference/cauchy_combined.html +https://lcrawlab.github.io/mvMAPIT/reference/fishers_combined.html +https://lcrawlab.github.io/mvMAPIT/reference/harmonic_combined.html +https://lcrawlab.github.io/mvMAPIT/reference/index.html +https://lcrawlab.github.io/mvMAPIT/reference/mvmapit.html +https://lcrawlab.github.io/mvMAPIT/reference/mvmapit_data.html +https://lcrawlab.github.io/mvMAPIT/reference/phillips_data.html +https://lcrawlab.github.io/mvMAPIT/reference/simulate_traits.html +https://lcrawlab.github.io/mvMAPIT/reference/simulated_data.html +