Enumerating transcription factors

Given the importance of the topic, it is not surprising that a number of bioinformatic research groups have published catalogs of transcription factors along with metadata about their features. Standard nomenclature for TFs has yet to be established. Gene symbols, motif sequences, and position-weight matrix catalog entries have all been used as TF identifiers.

In TFutils we have gathered information from four widely used resources, focusing specifically on human TFs: Gene Ontology (GO, Ashburner et al. (2000), in which GO:0003700 is the tag for the molecular function concept “DNA binding transcription factor activity”), CISBP (Weirauch et al. (2014)), HOCOMOCO (Kulakovskiy et al. (2018)), and the “c3 TFT (transcription factor target)” signature set of MSigDb (Subramanian et al. (2005)). Figure @ref(fig:lkupset) depicts the sizes of these catalogs, measured using counts of unique HGNC gene symbols. The enumeration for GO uses Bioconductor’s org.Hs.eg.db package to find direct associations from GO:0003700 to HGNC symbols. The enumeration for MSigDb is heuristic and involves parsing the gene set identifiers used in MSigDb for exact or close matches to HGNC symbols. For CISBP and HOCOMOCO, the associated web servers provide easily parsed tabular catalogs.

Sizes of TF catalogs and of intersections based on HGNC symbols for TFs.

Classification of transcription factors

As noted by Weirauch et al. (2014), interpretation of the “function and evolution of DNA sequences” is dependent on the analysis of sequence-specific DNA binding domains. These domains are dynamic and cell-type specific (Gertz et al. (2013)). Classifying TFs according to features of the binding domain is an ongoing process of increasing intricacy. Figure @ref(fig:TFclass) shows excerpts of hierarchies of terms related to TF type derived from GO (on the left) and TFclass (Wingender et al. (2018)). There is a disagreement between our enumeration of TFs based on GO in Figure @ref(fig:lkupset) and the 1919 shown in AmiGO, as the latter includes a broader collection of receptor activities.

Screenshots of AmiGO and TFClass hierarchy excerpts.

Table @ref(tab:classtab) provides examples of frequently encountered TF classifications in the CISBP and HOCOMOCO catalogs. The numerical components of the HOCOMOCO classes correspond to TFClass subfamilies (Wingender et al. (2018)).

Enumerating TF targets

The Broad Institute MSigDb (Subramanian et al. (2005)) includes a gene set collection devoted to cataloging TF targets. We have used Bioconductor’s GSEABase package to import and serialize the gmt representation of this collection.

TFutils::tftColl

## GeneSetCollection
##   names: AAANWWTGC_UNKNOWN, AAAYRNCTG_UNKNOWN, ..., GCCATNTTG_YY1_Q6 (615 total)
##   unique identifiers: 4208, 481, ..., 56903 (12774 total)
##   types in collection:
##     geneIdType: EntrezIdentifier (1 total)
##     collectionType: NullCollection (1 total)

Names of TFs for which target sets are assembled are encoded in a systematic way, with underscores separating substrings describing motifs, genes, and versions. Some peculiarity in nomenclature in the MSigDb labels can be observed:

grep("NFK", names(TFutils::tftColl), value=TRUE)

## [1] "NFKAPPAB65_01"         "NFKAPPAB_01"           "NFKB_Q6"              
## [4] "NFKB_C"                "NFKB_Q6_01"            "GGGNNTTTCC_NFKB_Q6_01"

Manual curation will be needed to improve the precision with which MSigDb TF target sets can be associated with specific TFs or motifs.

Quantitative predictions of TF binding affinities

In this subsection we address representation of putative binding sites. First we illustrate how to represent sequence-based affinity measures and the binding site locations implied by these. We then discuss use of results of ChIP-seq experiments for cell-type-specific binding site enumeration.

Affinity scores based on reference sequence. The FIMO algorithm of the MEME suite (Grant, Bailey, and Noble (2011)) was used to score the human reference genome for TF binding affinity for 689 motif matrices to which genes are associated. Full details are provided in Sonawane et al. (2017). Sixteen (16) tabix-indexed BED files are lodged in an AWS S3 bucket for illustration purposes.

library(GenomicFiles)
data(fimo16)
fimo16

## GenomicFiles object with 0 ranges and 16 files: 
## files: M0635_1.02sort.bed.gz, M3433_1.02sort.bed.gz, ..., M6159_1.02sort.bed.gz, M6497_1.02sort.bed.gz 
## detail: use files(), rowRanges(), colData(), ...

head(colData(fimo16))

## DataFrame with 6 rows and 2 columns
##          Mtag        HGNC
##   <character> <character>
## 1     M0635_1      DMRTC2
## 2     M3433_1       HOXA3
## 3     M3467_1        IRF1
## 4     M3675_1      POU2F1
## 5     M3698_1        TP53
## 6     M3966_1       STAT1

We harvest scores in a genomic interval of interest (bound to fimo16 in the rowRanges assignment below) using reduceByFile. This yields a list with one element per file. Each such element holds a list of scanTabix results, one per query range.

library(BiocParallel)
register(SerialParam()) # important for macosx?
rowRanges(fimo16) = GRanges("chr17", IRanges(38.077e6, 38.084e6))
rr = GenomicFiles::reduceByFile(fimo16, MAP=function(r,f)
  scanTabix(f, param=r))

scanTabix produces a list of vectors of text strings, which we parse with data.table::fread. The resulting tables are then reduced to a genomic location and -log10 of the p-value derived from the binding affinity statistic of FIMO in the vicinity of that location.

asdf = function(x) data.table::fread(paste0(x, collapse="\n"), header=FALSE)
gg = lapply(rr, function(x) {
       tmp = asdf(x[[1]][[1]]) 
       data.frame(loc=tmp$V2, score=-log10(tmp$V7))
     })
for (i in 1:length(gg))  gg[[i]]$tf = colData(fimo16)[i,2]

It turns out there are too many distinct TFs to display names individually, so we label the scores with the names of the associated TF families as defined in CISBP.

matchcis = match(colData(fimo16)[,2], cisbpTFcat[,2])
famn = cisbpTFcat[matchcis,]$Family_Name
for (i in 1:length(gg))  gg[[i]]$tffam = famn[i]
nn = do.call(rbind, gg)

A simple display of predicted TF binding affinity near the gene ORMDL3 is provided in Figure @ref(fig:finish).

TF binding in the vicinity of gene ORMDL3. Points are -log10-transformed FIMO-based p-values colored according to TF class as annotated in CISBP. Segments at bottom of plot are transcribed regions of ORMDL3 according to UCSC gene models in build hg19.

TF binding predictions based on ChIP-seq data from ENCODE. The ENCODE project provides BED-formatted reports on ChIP-seq experiments for many combinations of cell type and DNA-binding factors. TFutils includes a table encode690 that gives information on 690 experiments involving pairs formed from 91 cell lines and 161 TFs for which results have been recorded as GRanges instances that can be acquired with the AnnotationHub package. Positional relationships between cell-type specific binding sites and genomic features can be investigated. An illustration is given in Figure @ref(fig:lkbi), in which is it suggested that in HepG2 cells, CEBPB exhibits a distinctive pattern of binding in the vicinity of ORMDL3.

Binding of CEBPB in the vicinity of ORMDL3 derived from ChIP-seq experiments in four cell lines reported by ENCODE. Colored rectangles at top are regions identified as narrow binding peaks, arrows in bottom half are exons in ORMDL3. Arrows sharing a common vertical position are members of the same transcript as cataloged in Ensembl version 75.

Summary

We have compared enumerations of human transcription factors by different projects, provided access to two forms of binding domain classification, and illustrated the use of cloud-resident genome-wide binding predictions. In the next section we review selected details of data structures and methods of the TFutils package.

Implementation

The TFutils package is designed to lower barriers to usage of key findings of TF biology in human genome research. TFutils is supplied as a conventional R package distributed with, and making use of, the Bioconductor software ecosystem. TFutils includes ready-to-use reference data, tools for visualizing binding sites, and tools that simplify integrative use of TF binding information with GWAS findings.

Infrastructure for interacting with components of TFutils

Interactive enumeration of TF targets implicated in GWAS. The TFtargs function runs a shiny app that permits selection of a TF in the nomenclature of the MSigDb c3/TFT gene set collection. The app will search an object provided by the gwascat package for references in the MAPPED_GENE field that match the targets of the selected TF. Figure @ref(fig:lktarapp) gives an illustration.

TFtargs() screenshot. This example reports on recent EBI GWAS catalog hits on chromosome 17 only.

The TFCatalog S4 class. Reference catalogs for TF biology are structured with the TFCatalog S4 class. Two essential components for managing a catalog are the native TF identifier for the catalog and the HGNC gene symbol typically used to name the TF. The TFCatalog class includes a name field to name the catalog, and a character vector with elements comprised of the native identifiers for catalogued TFs.
For example, CISBP uses T004843_1.02 to refer to motifs associated with gene TFAP2B. There are five such motifs, three derived from SELEX, one from Transfac, and one from Hocomoco.

A data.frame instance that has an obligatory column named ‘HGNC’ can include any collection of fields that offer metadata about the TF in the specified catalog. Here is how we construct and view a TFCatalog object using the CISBP reference data.

data(cisbpTFcat)
TFs_CISBP = TFCatalog(name="CISBP.info", 
   nativeIds=cisbpTFcat[,1],
   HGNCmap = cisbpTFcat)
TFs_CISBP

## TFutils TFCatalog instance CISBP.info 
##  7592 native Ids, including
##    T004843_1.02 ... T153733_1.02 
##  1551 unique HGNC tags, including
##    TFAP2B TFAP2B ... ZNF10 ZNF350

Operation: Use cases

In this section we consider applications of the tools in genetic epidemiology. First we look for TFs that may harbor variants associated with traits in the EBI GWAS catalog. Then we show how to enumerate traits associated with targets of a selected TF.

TFs that are direct GWAS hits for a given trait. directHitsInCISBP accepts a string naming a trait , and returns a data.frame of TFs identified as “mapped genes” for the trait, with their TF “family name”.

library(dplyr)
library(magrittr)
library(gwascat)
data(ebicat37)
directHitsInCISBP("Rheumatoid arthritis", ebicat37)

## Joining, by = "HGNC"

##      HGNC Family_Name
## 1  ARID5B ARID/BRIGHT
## 7   EOMES       T-box
## 15  GATA3        GATA
## 35  JAZF1     C2H2 ZF
## 37  MECP2         MBD
## 45   MTF1     C2H2 ZF
## 57    REL         Rel
## 65  STAT4        STAT
## 79   AIRE        SAND
## 82   IRF5         IRF

Traits mapped to genes that are targets of a given TF

topTraitsOfTargets will acquire the targets of a selected TF, check for hits in these genes in a given GWAS catalog instance, and tabulate the most commonly reported traits.

tt = topTraitsOfTargets("MTF1", TFutils::tftColl, ebicat37)

## remapping identifiers of input GeneSetCollection to Symbol...

## done

head(tt)

##                              DISEASE.TRAIT MAPPED_GENE       SNPS CHR_ID
## 1                        Atopic dermatitis        TNXB rs41268896      6
## 2                        Atopic dermatitis        TNXB rs12153855      6
## 3                        Atopic dermatitis       KIF3A  rs2897442      5
## 4 Attention deficit hyperactivity disorder      SEMA3A   rs797820      7
## 5 Attention deficit hyperactivity disorder        DNM1  rs2502731      9
## 6 Attention deficit hyperactivity disorder        GPC6  rs7995215     13
##     CHR_POS
## 1  32102292
## 2  32107027
## 3 132713335
## 4  83979723
## 5 128214278
## 6  93756253

table(tt[,1])

## 
##                        Atopic dermatitis 
##                                        3 
## Attention deficit hyperactivity disorder 
##                                        3 
##                                   Height 
##                                        7 
##                  Menarche (age at onset) 
##                                        4 
##                   Obesity-related traits 
##                                       11 
##                     Rheumatoid arthritis 
##                                        3