library(LTFGRS)
library(igraph)
library(dplyr)

We will here demonstrate some basic data management operations using igraph objects created with the prepare_graph() function in LTFGRS.

# hand curated trio information, taken from LTFHPlus vignette:
# https://emilmip.github.io/LTFHPlus/articles/FromTrioToFamilies.html
family = tribble(
  ~id, ~momcol, ~dadcol,
  "pid", "mom", "dad",
  "sib", "mom", "dad",
  "mhs", "mom", "dad2",
  "phs", "mom2", "dad",
  "mom", "mgm", "mgf",
  "dad", "pgm", "pgf",
  "dad2", "pgm2", "pgf2",
  "paunt", "pgm", "pgf",
  "pacousin", "paunt", "pauntH",
  "hspaunt", "pgm", "newpgf",
  "hspacousin", "hspaunt", "hspauntH",
  "puncle", "pgm", "pgf",
  "pucousin", "puncleW", "puncle",
  "maunt", "mgm", "mgf",
  "macousin", "maunt", "mauntH",
  "hsmuncle", "newmgm", "mgf",
  "hsmucousin", "hsmuncleW", "hsmuncle"
)

# creating a graph for the family
graph = prepare_graph(.tbl = family, icol = "id", mcol = "momcol", fcol = "dadcol")

Below, we print an igraph object. For additional details on the print of an igraph object, please see this documentation. It contains some summary information about the graph:

print(graph)

## IGRAPH 350dfc6 DN-- 31 44 -- 
## + attr: name (v/c)
## + edges from 350dfc6 (vertex names):
##  [1] dad     ->pid        mom     ->pid        dad     ->sib       
##  [4] mom     ->sib        dad2    ->mhs        mom     ->mhs       
##  [7] dad     ->phs        mom2    ->phs        mgf     ->mom       
## [10] mgm     ->mom        pgf     ->dad        pgm     ->dad       
## [13] pgf2    ->dad2       pgm2    ->dad2       pgf     ->paunt     
## [16] pgm     ->paunt      pauntH  ->pacousin   paunt   ->pacousin  
## [19] newpgf  ->hspaunt    pgm     ->hspaunt    hspauntH->hspacousin
## [22] hspaunt ->hspacousin pgf     ->puncle     pgm     ->puncle    
## + ... omitted several edges

The first row contains “DN– 31 44 –”. This means that the graph is directed (D) and named (N), meaning the vertices are named. The graph contains 31 vertices (individuals in the graph) and 44 edges (familial links). However, the construction of the graph in prepare_graph() adds dummy connections between siblings to assist with relatedness calculations and to ensure neighbourhood graphs will contain all relevant individuals. The dummy relations can be removed with the following command, which removes all undirected edges and the total number can be read from the print:

delete_edges(graph, E(graph)[which_mutual(graph)])

## IGRAPH 71324d5 DN-- 31 34 -- 
## + attr: name (v/c)
## + edges from 71324d5 (vertex names):
##  [1] dad     ->pid        mom     ->pid        dad     ->sib       
##  [4] mom     ->sib        dad2    ->mhs        mom     ->mhs       
##  [7] dad     ->phs        mom2    ->phs        mgf     ->mom       
## [10] mgm     ->mom        pgf     ->dad        pgm     ->dad       
## [13] pgf2    ->dad2       pgm2    ->dad2       pgf     ->paunt     
## [16] pgm     ->paunt      pauntH  ->pacousin   paunt   ->pacousin  
## [19] newpgf  ->hspaunt    pgm     ->hspaunt    hspauntH->hspacousin
## [22] hspaunt ->hspacousin pgf     ->puncle     pgm     ->puncle    
## + ... omitted several edges

Hence, there are 34 true familial relations in the graph. The number of vertices can also be extracted directly with vcount(graph) and the number of edges can be extracted with ecount(graph).

The second row is + attr: name (v/c), which indicates that the graph has an attribute, i.e. a piece of information attached to each vertex (v). The attribute is called name and is of type character (c). The name attribute contains the individual IDs. The third row onwards list (a subset of) the edges in the graph.

An igraph object is a hidden list structure. To see the full structure of the igraph object, the str() function can be used:

str(graph)

## Class 'igraph'  hidden list of 10
##  $ : num 31
##  $ : logi TRUE
##  $ : num [1:44] 0 1 0 1 2 1 0 3 4 5 ...
##  $ : num [1:44] 23 23 22 22 24 24 25 25 1 1 ...
##  $ : NULL
##  $ : NULL
##  $ : NULL
##  $ : NULL
##  $ :List of 4
##   ..$ : num [1:3] 1 0 1
##   ..$ : Named list()
##   ..$ :List of 1
##   .. ..$ name: chr [1:31] "dad" "mom" "dad2" "mom2" ...
##   ..$ : Named list()
##  $ :<environment: 0x55ddeedf4d70>

Manupulations of the graph can therefore be done with suitable list operations, however, igraph also contains a number of helper functions to make common graph operations easier.

Manipulating attributes

Below is an example of how an attribute can be added to a vertex in an igraph object:

graph = set_vertex_attr(graph = graph, # graph to add attribute to
                        name  = "status",  # name of attribute to be added
                        index = V(graph), # which vertices to add attribute to, V(graph) adds to all vertices
                        value = sample(c(FALSE, TRUE), size = vcount(graph), replace = TRUE, prob = c(0.95, 0.05))) # value of the attiribute to be added
print(graph)

## IGRAPH 350dfc6 DN-- 31 44 -- 
## + attr: name (v/c), status (v/l)
## + edges from 350dfc6 (vertex names):
##  [1] dad     ->pid        mom     ->pid        dad     ->sib       
##  [4] mom     ->sib        dad2    ->mhs        mom     ->mhs       
##  [7] dad     ->phs        mom2    ->phs        mgf     ->mom       
## [10] mgm     ->mom        pgf     ->dad        pgm     ->dad       
## [13] pgf2    ->dad2       pgm2    ->dad2       pgf     ->paunt     
## [16] pgm     ->paunt      pauntH  ->pacousin   paunt   ->pacousin  
## [19] newpgf  ->hspaunt    pgm     ->hspaunt    hspauntH->hspacousin
## [22] hspaunt ->hspacousin pgf     ->puncle     pgm     ->puncle    
## + ... omitted several edges

Notice that the status attribute has been added to the print of the graph above. The attribute is a vector (v) of type logical (l) and contains either FALSE or TRUE.

All attributes can be extracted with the vertex_attr() function, since attributes can be any type of data format it is stored as a list:

vertex_attr(graph)

## $name
##  [1] "dad"        "mom"        "dad2"       "mom2"       "mgf"       
##  [6] "mgm"        "pgf"        "pgm"        "pgf2"       "pgm2"      
## [11] "pauntH"     "paunt"      "newpgf"     "hspauntH"   "hspaunt"   
## [16] "puncle"     "puncleW"    "mauntH"     "maunt"      "newmgm"    
## [21] "hsmuncle"   "hsmuncleW"  "sib"        "pid"        "mhs"       
## [26] "phs"        "pacousin"   "hspacousin" "pucousin"   "macousin"  
## [31] "hsmucousin"
## 
## $status
##  [1] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [13] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [25] FALSE FALSE FALSE FALSE FALSE FALSE FALSE

Attributes can also be added in a list-style assignment:

# adding another vertex attribute with a list-style assignment
V(graph)$age = sample(20:80, size = vcount(graph), replace = TRUE)
vertex_attr(graph)

## $name
##  [1] "dad"        "mom"        "dad2"       "mom2"       "mgf"       
##  [6] "mgm"        "pgf"        "pgm"        "pgf2"       "pgm2"      
## [11] "pauntH"     "paunt"      "newpgf"     "hspauntH"   "hspaunt"   
## [16] "puncle"     "puncleW"    "mauntH"     "maunt"      "newmgm"    
## [21] "hsmuncle"   "hsmuncleW"  "sib"        "pid"        "mhs"       
## [26] "phs"        "pacousin"   "hspacousin" "pucousin"   "macousin"  
## [31] "hsmucousin"
## 
## $status
##  [1] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [13] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [25] FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## 
## $age
##  [1] 34 78 51 68 74 70 69 51 49 25 34 38 63 80 27 49 21 69 48 69 43 66 37 76 78
## [26] 54 79 25 79 59 49

If an attribute is no longer needed, it can be removed with the delete_vertex_attr() function:

delete_vertex_attr(graph = graph, name = "age") %>% # age is not removed from graph, since it is not stored.
  vertex_attr()

## $name
##  [1] "dad"        "mom"        "dad2"       "mom2"       "mgf"       
##  [6] "mgm"        "pgf"        "pgm"        "pgf2"       "pgm2"      
## [11] "pauntH"     "paunt"      "newpgf"     "hspauntH"   "hspaunt"   
## [16] "puncle"     "puncleW"    "mauntH"     "maunt"      "newmgm"    
## [21] "hsmuncle"   "hsmuncleW"  "sib"        "pid"        "mhs"       
## [26] "phs"        "pacousin"   "hspacousin" "pucousin"   "macousin"  
## [31] "hsmucousin"
## 
## $status
##  [1] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [13] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [25] FALSE FALSE FALSE FALSE FALSE FALSE FALSE

Using attributes for calculations

First, we create family graphs for all probands in the population graph created above. Here, we set ndegree = 1, meaning that only first-degree relatives (parents, siblings, children) will be included in the family graphs.

fam_graphs = get_family_graphs(pop_graph = graph, ndegree = 1, proband_vec = V(graph)$name)

Below is a function that takes a proband ID and a family graph as input arguments. The function is intended to calculate a logical variable indicating whether any family member in the (neighobourhood) graph is diagnosed with a phenotype of interest. This is equivalent to calculating a binary family history indicator based on all identified family members.

get_FH = function(cur_proband, cur_graph) {
  fam_indx = which(V(cur_graph)$name != cur_proband)
  any(vertex_attr(cur_graph, name = "status", index = V(cur_graph)$name[fam_indx]))
}

We can then use this function to calculate family history for all probands in the fam_graphs tibble created above:

fam_graphs %>% 
  mutate(FH = purrr::map2_lgl(.x = fid, .y = fam_graph,
                         .f = ~ get_FH(cur_proband = .x, cur_graph = .y)))

## # A tibble: 31 × 3
##    fid   fam_graph FH   
##    <chr> <list>    <lgl>
##  1 dad   <igraph>  FALSE
##  2 mom   <igraph>  FALSE
##  3 dad2  <igraph>  FALSE
##  4 mom2  <igraph>  FALSE
##  5 mgf   <igraph>  FALSE
##  6 mgm   <igraph>  FALSE
##  7 pgf   <igraph>  FALSE
##  8 pgm   <igraph>  FALSE
##  9 pgf2  <igraph>  FALSE
## 10 pgm2  <igraph>  FALSE
## # ℹ 21 more rows

Removing vertices with certain attribute values

We can also remove vertices based on attribute values. For example, to remove all individuals older than 50 years from the graph above, we can do:

delete_vertices(graph = graph,
                v = V(graph)[which(vertex_attr(graph, name = "age") > 50)])

## IGRAPH b8091e5 DN-- 13 9 -- 
## + attr: name (v/c), status (v/l), age (v/n)
## + edges from b8091e5 (vertex names):
## [1] dad     ->sib        hspaunt ->hspacousin hsmuncle->hsmucousin
## [4] paunt   ->dad        puncle  ->dad        dad     ->paunt     
## [7] puncle  ->paunt      dad     ->puncle     paunt   ->puncle

- Manipulating attributes
  - Using attributes for calculations
  - Removing vertices with certain attribute values

iGraph data management

Manipulating attributes

Using attributes for calculations

Removing vertices with certain attribute values