Contents

1 Introduction

This document explores different ways to assess the relevance of a gene for a set of human phenotypes using the PCAN package.

library(PCAN)
## Loading required package: BiocParallel
## Some functions can be parallelized.
## They use the bpmapply function from the BiocParallel library.
## Follow instructions provided in the BiocParallel manual to
## configure your parallelization backend.
## options(MulticoreParam=quote(MulticoreParam(workers=4)))

1.1 Use case

This document demonstrates the capabilities of the PCAN package through the analysis of the following example.

Here, we pretend that we don’t know anything about the genetics of the Joubert syndrome 9. The symptoms of this syndrome can be formally described by the following terms from the Human Phenotype Ontology (Köhler et al. 2014) (according to this version of the phenotype_annotation.tab):

HP Name
HP:0000483 Astigmatism
HP:0000510 Retinitis pigmentosa
HP:0000518 Cataract
HP:0000639 Nystagmus
HP:0001249 Intellectual disability
HP:0001250 Seizures
HP:0002119 Ventriculomegaly
HP:0002419 Molar tooth sign on MRI

Let’s store these phenoytpes in the hpOfInterest vector:

hpOfInterest <- c("HP:0000483", "HP:0000510", "HP:0000518", "HP:0000639", "HP:0001249", "HP:0001250", "HP:0002119", "HP:0002419")

The CC2D2A is known to be associated to the Joubert syndrome 9 according to OMIM. Let’s pretend in the frame of this example that we don’t know this association and that this gene came out from sequencing data related to an individual suffering from the Joubert syndrome 9.

The aim of the following analyses is to assess the relevance of this gene for the set of phenotypes under focus.

1.2 Prior knowledge

To achieve the goal described above, we rely on prior knowledge about genetics of disorders and human phenotypes.

1.2.1 Mendelian disorders and associated genes

Genes known to be associated to disorders were identified using clinVar (Landrum et al. 2014). In this package we provide part of this information taken from the ClinVarFullRelease_2015-05.xml.gz file.

The geneByTrait data frame provides the entrez gene IDs associated to disorder. Association were filtered according to the following criteria:

  • they all come from OMIM;
  • they all have a pathogenic clinical status;
  • they all have one of the following origins: germline, de novo, inherited, maternal, paternal, biparental or uniparental.
data(geneByTrait, package="PCAN")
head(geneByTrait, n=3)
##       entrez   db     id
## 2       3293 OMIM 264300
## 12      8813 OMIM 608799
## 22 101927631 OMIM 608799
dim(geneByTrait)
## [1] 4569    3

The traitDef data frame provides the names of the different disorders:

data(traitDef, package="PCAN")
head(traitDef, n=3)
##        db     id                                 name
## 8015 OMIM 610443      17q21.31 microdeletion syndrome
## 8016 OMIM 616034 2,4-Dienoyl-CoA reductase deficiency
## 8017 OMIM 300438   2-methyl-3-hydroxybutyric aciduria

The geneDef data frame provides basic information about the genes:

data(geneDef, package="PCAN")
head(geneDef, n=3)
##   entrez                     name symbol
## 1   2720    galactosidase, beta 1   GLB1
## 2   1509              cathepsin D   CTSD
## 3    846 calcium-sensing receptor   CASR

Since we pretend that we don’t know the association between the Joubert syndrome 9 (disId <- "612285") and CC2D2A (genId <- "57545"), let’s remove it from the geneByTrait data frame:

geneByTrait <- geneByTrait[
    which(geneByTrait$id!=disId | geneByTrait$entrez!=genId),
]
dim(geneByTrait)
## [1] 4568    3

1.2.2 Human phenotype of mendelian disorders

OMIM disorders were associated to human phenotype using the phenotype_annotation.tab file. The associations are available in the hpByTrait data frame:

data(hpByTrait, package="PCAN")
head(hpByTrait, n=3)
##               hp   db     id
## 71837 HP:0000003 OMIM 100100
## 71839 HP:0000010 OMIM 100100
## 71840 HP:0000028 OMIM 100100

Description of the different HP terms were obtained from the hp.obo file. They are available in the hpDef data frame:

data(hpDef, package="PCAN")
head(hpDef, n=3)
##           id                                     name
## 2 HP:0000002               Abnormality of body height
## 3 HP:0000003             Multicystic kidney dysplasia
## 7 HP:0000008 Abnormality of female internal genitalia

The same hp.obo file was used to get the descendant HP and the ancestor HP for each HP term. They are available in the hp_descendants and the hp_ancestors lists respectively:

data(hp_descendants, hp_ancestors, package="PCAN")
lapply(head(hp_descendants, n=3), head)
## $`HP:0000002`
## [1] "HP:0000098" "HP:0000839" "HP:0001519" "HP:0001533" "HP:0001548"
## [6] "HP:0003498"
## 
## $`HP:0000003`
## [1] "HP:0000003"
## 
## $`HP:0000008`
## [1] "HP:0000013" "HP:0000130" "HP:0000131" "HP:0000132" "HP:0000134"
## [6] "HP:0000136"
lapply(head(hp_ancestors, n=3), head)
## $`HP:0000002`
## [1] "HP:0001507" "HP:0000118" "HP:0000001" "HP:0000002"
## 
## $`HP:0000003`
## [1] "HP:0000107" "HP:0012210" "HP:0000077" "HP:0010935" "HP:0000079"
## [6] "HP:0000119"
## 
## $`HP:0000008`
## [1] "HP:0000812" "HP:0010460" "HP:0012243" "HP:0000078" "HP:0000119"
## [6] "HP:0000118"

We only kept information related to the descendants of Phenotypic abnormality (HP:0000118).

The geneByHp data frame, showing gene associated to each HP term, has been created from the geneByTrait and the hpByTraitdata frames. This data frame is available in the package: data(geneByHp, package="PCAN"). However, since we pretend in the frame of this example that we don’t know the association between the Joubert syndrome 9 and CC2D2A, we need to rebuild the geneByHp data frame using our filtered geneByTrait data frame:

geneByHp <- unique(merge(
    geneByTrait,
    hpByTrait,
    by="id"
)[,c("entrez", "hp")])
head(geneByHp, n=3)
##   entrez         hp
## 1   1131 HP:0005199
## 2   1131 HP:0001374
## 3   1131 HP:0003422

1.2.3 Custom prior knowledge

Several objects representing some biological knowledge are attached to this package for the convenience of the user. Nevertheless, the package functions could be used with other source of knowledge depending on the user needs and the updates of the different resources.

In the frame of this project we provide some R scripts (available in the inst/DataPackage-Generator/ folder of the package) allowing the generation of a package gathering up-to-date information from different databases:

Generating such a package is a way to get up-to-date information from these sources. However, some other means could be more convenient for some users. That’s why this data resource is not tightly coupled with the PCAN package.

2 Direct comparison of phenotypes

2.1 Phenotypes associated to the gene candidate

Let’s use our prior knowledge to find disorders, and eventually human phenotypes, associated to the gene candidate CC2D2A:

genDis <- traitDef[
    match(
        geneByTrait[which(geneByTrait$entrez==genId), "id"],
        traitDef$id
    ),
]
genDis
##         db     id                   name
## 10479 OMIM 249000 Meckel syndrome type 1
## 10484 OMIM 612284 Meckel syndrome type 6
## 8809  OMIM 216360         COACH syndrome
genHpDef <- hpDef[
    match(
        geneByHp[which(geneByHp$entrez==genId), "hp"],
        hpDef$id
    ),
]
genHp <- genHpDef$id
dim(genHpDef)
## [1] 119   2
head(genHpDef)
##              id                              name
## 1765 HP:0002342 Intellectual disability, moderate
## 1974 HP:0002650                         Scoliosis
## 136  HP:0000154                        Wide mouth
## 1696 HP:0002240                      Hepatomegaly
## 2157 HP:0002896             Neoplasm of the liver
## 1716 HP:0002269 Abnormality of neuronal migration

How many of these 119 HP terms are shared with the 8 HP related to the syndrome under focus?

genHpDef[which(genHpDef$id %in% hpOfInterest),]
##              id                    name
## 506  HP:0000639               Nystagmus
## 1817 HP:0002419 Molar tooth sign on MRI
## 972  HP:0001250                Seizures

However some of the 116 CC2D2A associated HPs which are not among the 8 HPs of interest present strong similarity with the last ones. For example the Dandy-Walker malformation (HP:0001305) phenotype, associated to CC2D2A, is an indirect but closely related descendant of Ventriculomegaly (HP:0002119) phenotype of interest.

The following steps describe a way to measure similarity between different HP terms.

2.2 Information content and semantic similariy

Here we measure semantic similarity between HPs using gene information content (IC). The formula below shows how information content is computed for each HP term \(p\):

\[ IC_{p} = -ln\left(\frac{|p|}{|root|}\right) \]

Where \(|p|\) is the number of gene associated to the HP term \(p\) and all its descendants. \(root\), in our case, is Phenotypic abnormality (HP:0000118). By definition: \(IC_{root} = -ln\left(\frac{|root|}{|root|}\right) = 0\).

Let’s use the computeHpIC function to compute IC for all HP terms descendants of Phenotypic abnormality in the human phenotype ontology. This function needs to know the genes associated to each HP and the descendants of each HP term.

info <- unstack(geneByHp, entrez~hp)
ic <- computeHpIC(info, hp_descendants)

Let’s have a look at the distribution of IC:

IC is an measure of the specificity of genes associated to HPs. The higher IC, the more specific.

Semantic similarity (\(SS_{p_{1}p_{2}}\)) between two HP terms is then defined as the IC of the most informative common ancestor (MICA) (i.e. showing the higher IC).

Let’s use the clacHpSim function to compute the semantic similarity between different couples of HP terms:

hp1 <- "HP:0000518"
hp2 <- "HP:0030084"
hpDef[which(hpDef$id %in% c(hp1, hp2)), 1:2]
##              id         name
## 401  HP:0000518     Cataract
## 9645 HP:0030084 Clinodactyly
calcHpSim(hp1, hp2, IC=ic, ancestors=hp_ancestors)
## [1] 0
hp1 <- "HP:0002119"
hp2 <- "HP:0001305"
hpDef[which(hpDef$id %in% c(hp1, hp2)), 1:2]
##              id                      name
## 1012 HP:0001305 Dandy-Walker malformation
## 1601 HP:0002119          Ventriculomegaly
calcHpSim(hp1, hp2, IC=ic, ancestors=hp_ancestors)
## [1] 2.850015

2.3 Comparing two sets of phenotype

Now, we can compute semantic similarity between all HP of interest and CC2D2A associated HPs using the compMat function:

compMat <- compareHPSets(
    hpSet1=genHp, hpSet2=hpOfInterest,
    IC=ic,
    ancestors=hp_ancestors,
    method="Resnik",
    BPPARAM=SerialParam()
)
dim(compMat)
## [1] 119   8
head(compMat)
##            HP:0000483 HP:0000510 HP:0000518 HP:0000639 HP:0001249 HP:0001250
## HP:0002342          0          0          0          0  1.4502253  0.5169219
## HP:0002650          0          0          0          0  0.0000000  0.0000000
## HP:0000154          0          0          0          0  0.0000000  0.0000000
## HP:0002240          0          0          0          0  0.0000000  0.0000000
## HP:0002896          0          0          0          0  0.0000000  0.0000000
## HP:0002269          0          0          0          0  0.4262711  0.4262711
##            HP:0002119 HP:0002419
## HP:0002342  0.4262711  0.4262711
## HP:0002650  0.0000000  0.0000000
## HP:0000154  0.0000000  0.0000000
## HP:0002240  0.0000000  0.0000000
## HP:0002896  0.0000000  0.0000000
## HP:0002269  0.9002305  0.9002305

Then we compute the symmetric semantic similarity score of the matrix to get single value corresponding to similarity between the two sets of HP terms: the HP terms of interest and CC2D2A associated HPs.

hpSetCompSummary(compMat, method="bma", direction="symSim")
## [1] 1.357548

Unfortunately it is not easy to interpret such a score and to assess it’s significance. To do it we need to compare the score of the candidate gene (CC2D2A) to the score of all the other genes for which we can compute it. Let’s compute the score for all the genes:

## Compute semantic similarity between HP of interest and all HP terms
## This step is time consumming and can be parallelized.
## Use the BPPARAM parameter to specify your own 
## back-end with appropriate number of workers.
hpGeneResnik <- compareHPSets(
    hpSet1=names(ic), hpSet2=hpOfInterest,
    IC=ic,
    ancestors=hp_ancestors,
    method="Resnik",
    BPPARAM=SerialParam()
)
## Group the results by gene
hpByGene <- unstack(geneByHp, hp~entrez)
hpMatByGene <- lapply(
    hpByGene,
    function(x){
        hpGeneResnik[x, , drop=FALSE]
    }
)
## Compute the corresponding scores
resnSss <- unlist(lapply(
    hpMatByGene,
    hpSetCompSummary,
    method="bma", direction="symSim"
))
## Get the score of the gene candidate
candScore <- resnSss[genId]
candScore
##    57545 
## 1.357548

And now, we can compare the score of the candidate to all the others:

candRank <- sum(resnSss >= candScore)
candRank
## [1] 130
candRelRank <- candRank/length(resnSss)
candRelRank
## [1] 0.04086765

According to a direct comparison, the candidate gene CC2D2A is in the top 4.1% genes the most relevant for the set of HPs of interest. This result can be used for candidate prioritization.

3 The pathway consensus approach

Often, gene candidates are not known yet to be associated to any genetic disorders. In such cases the prior knowledge can not be used to associate HP terms to the gene and the direct comparison of HP sets is not possible. In such situation we can focus genes known to interact with the gene of interest or known to be involved in the same biological processes and compute a consensus score taking all of them into account. This pathway consensus approach can also be used in addition to the direct comparison to provide further confidence or insight into the relationship between the gene candidate and the syndrome under focus.

3.1 Additional prior knowledge

To be able to apply such an approach we obviously need some information about gene pathways or gene network. For the convenience of the user we provide such an information within the package. However the user can use any kind of resource depending on the needs.

Gene belonging to Reactome (Croft et al. 2014) pathways are provided in the hsEntrezByRPath object. The name of the pathway can be found in the rPath data frame.

data(hsEntrezByRPath, rPath, package="PCAN")
head(rPath, n=3)
##                   Pathway
## REACT_268024 REACT_268024
## REACT_75842   REACT_75842
## REACT_116145 REACT_116145
##                                                             Pathway_name
## REACT_268024                                    Intraflagellar transport
## REACT_75842  Antigen processing: Ubiquitination & Proteasome degradation
## REACT_116145                             PPARA activates gene expression
lapply(head(hsEntrezByRPath, 3), head)
## $REACT_268024
## [1] "23059" "51626" "83658" "83657" "79659" "8655" 
## 
## $REACT_75842
## [1] "29882" "10393" "51529" "25847" "64682" "29945"
## 
## $REACT_116145
## [1] "19"    "34"    "2180"  "51129" "183"   "27063"

The STRING database (Jensen et al. 2009) was used to get gene interactions. This information, focused on Homo sapiens genes and on interaction with a score higher than 500, can be found in the hqStrNw data frame:

data(hqStrNw, package="PCAN")
head(hqStrNw, n=3)
##   gene1 gene2 upstream
## 1   381   381    FALSE
## 2   381 23647    FALSE
## 3   381  8546    FALSE

3.2 Genes belonging to the pathways of the candidate

Here we are going to assess the relevance of genes involved in the same pathways as CC2D2A for the HP terms of interest.

First, let’s identify the pathways in which CC2D2A is involved:

candPath <- names(hsEntrezByRPath)[which(unlist(lapply(
    hsEntrezByRPath,
    function(x) genId %in% x
)))]
rPath[which(rPath$Pathway %in% candPath),]
##                   Pathway                                       Pathway_name
## REACT_267965 REACT_267965 Anchoring of the basal body to the plasma membrane

Then we can retrieve the symmetric semantic similarity scores for all these genes when the information is available. Let’s use the hpGeneListComp function:

rPathRes <- hpGeneListComp(
    geneList=hsEntrezByRPath[[candPath]],
    ssMatByGene = hpMatByGene,
    geneSSScore = resnSss
)

This function returns a list with many information. Have a look at ?hpGeneListComp to get a complete description of this output. Among the scores element of this output provides the scores for the genes in the submitted list:

length(rPathRes$scores)
## [1] 88
sum(!is.na(rPathRes$scores))
## [1] 39

Among the 88 genes belonging to the same pathway as CC2D2A, a score could be computed for only 39 of them.

The p.value element of the output provides the p-value returned by wilcox.test comparing these scores to the scores of all the genes not in the provided list.

This result show that in general the genes belonging to the Anchoring of the basal body to the plasma membrane pathway in which CC2D2A is involved are relevant for the set of phenotype of interest.

To get further insight we can explore the score of all the genes belonging to this pathway:

pathSss <- rPathRes$scores[which(!is.na(rPathRes$scores))]
names(pathSss) <- geneDef[match(names(pathSss), geneDef$entrez), "symbol"]
par(mar=c(7.1, 4.1, 4.1, 2.1))
barplot(
    sort(pathSss),
    las=2,
    ylab=expression(Sim[sym]),
    main=rPath[which(rPath$Pathway %in% candPath),"Pathway_name"]
)
p <- c(0.25, 0.5, 0.75, 0.95)
abline(
    h=quantile(resnSss, probs=p),
    col="#BE0000",
    lty=c(2, 1, 2, 2),
    lwd=c(2, 2, 2, 1)
)
text(
    rep(0,4),
    quantile(resnSss, probs=p),
    p,
    pos=3,
    offset=0,
    col="#BE0000",
    cex=0.6
)
legend(
    "topleft",
    paste0(
        "Quantiles of the distribution of symmetric semantic similarity\n",
        "scores for all the genes."
    ),
    lty=1,
    col="#BE0000",
    cex=0.6
)