In vjcitn/CSHstats: discussions and exercises on classical statistics for CSHL 2022

suppressPackageStartupMessages({
suppressMessages({
library(hca)
library(DT)
library(rols)
library(SingleCellExperiment)
library(ontoProc)
})
})

Road map

• Bioconductor reads a paper
• ontoProc and rols help with ontologies
• shiny helps communicate

A paper on the atlas of the prostate

The hca package

Surveying projects of the HCA

library(hca)
p = projects(size = 200); p = dplyr::bind_rows(p, hca_next(p)) # workaround bug upstream to 1.4.0
library(DT)
datatable(as.data.frame(p))

Picking a project; Enumerating and downloading loom files

projectId = "53c53cd4-8127-4e12-bc7f-8fe1610a715c"
file_filter <- filters(
    projectId = list(is = projectId),
    fileFormat = list(is = "loom")
)
pfile = files(file_filter)
pfile$projectTitle[1]
#pfile |> files_download()

Working with loom

Very superficial filtering (to 60000 cells) and development of PCA

library(LoomExperiment)
f1 = import("/home/stvjc/.cache/R/hca/36e582f7c6e_36e582f7c6e.loom")
f1
names(colData(f1))
library(scater)
sf1 = as(f1, "SingleCellExperiment")
sf1
library(scuttle)
assay(sf1[1:4,1:4])
assayNames(sf1) = "counts"
litsf1 = sf1[,1:60000]
z = DelayedArray::rowSums(assay(litsf1))
mean(z==0)
todrop = which(z==0)
litsf2 = litsf1[-todrop,]
assay(litsf2)
litsf2 = logNormCounts(litsf2)
litsf2 = runPCA(litsf2)

Whatchaget:

library(SingleCellExperiment)
if (!exists("litsf2")) load("litsf2.rda") # run code above, must have HDF5 in cache
metadata(litsf2)

> str(litsf2) # 22MB on disk (no quantifications)
Formal class 'DelayedMatrix' [package "DelayedArray"] with 1 slot
  ..@ seed:Formal class 'DelayedAperm' [package "DelayedArray"] with 2 slots
  .. .. ..@ perm: int [1:2] 2 1
  .. .. ..@ seed:Formal class 'DelayedSubset' [package "DelayedArray"] with 2 slots
  .. .. .. .. ..@ index:List of 2
  .. .. .. .. .. ..$ : int [1:60000] 1 2 3 4 5 6 7 8 9 10 ...
  .. .. .. .. .. ..$ : int [1:23420] 13 20 22 23 31 33 34 35 36 37 ...
  .. .. .. .. ..@ seed :Formal class 'HDF5ArraySeed' [package "HDF5Array"] with 7 slots
  .. .. .. .. .. .. ..@ filepath : chr "/home/stvjc/.cache/R/hca/36e582f7c6e_36e582f7c6e.loom"
  .. .. .. .. .. .. ..@ name     : chr "/matrix"
  .. .. .. .. .. .. ..@ as_sparse: logi FALSE
  .. .. .. .. .. .. ..@ type     : chr NA
  .. .. .. .. .. .. ..@ dim      : int [1:2] 382197 58347
  .. .. .. .. .. .. ..@ chunkdim : int [1:2] 64 64
  .. .. .. .. .. .. ..@ first_val: int 0

stvjc@stvjc-XPS-13-9300:~/CSAMA_HCA$ ls -tl /home/stvjc/.cache/R/hca/36e582f7c6e_36e582f7c6e.loom
-rw-rw-r-- 1 stvjc stvjc 1206062245 Jun 21 22:37 /home/stvjc/.cache/R/hca/36e582f7c6e_36e582f7c6e.loom

Working with iSEE

• Question: where is the "stop/exit" button?
• Question: can we embed iSEE (or components) in a vignette? Or is there an iSEE server?

Upshots

• easy to survey HCA with hca package
• easy to get experiments, metadata, quantifications for projects of interest
• iSEE really accelerates exploration and elaboration of data and claims

Ontologies, EBI OLS, rols (thanks Laurent Gatto!), ontoProc::ctmarks

Definition: from Wikipedia

In computer science and information science, an ontology encompasses a representation, formal naming, and definition of the categories, properties, and relations between the concepts, data, and entities that substantiate one, many, or all domains of discourse. More simply, an ontology is a way of showing the properties of a subject area and how they are related, by defining a set of concepts and categories that represent the subject.

Every academic discipline or field creates ontologies to limit complexity and organize data into information and knowledge. Each uses ontological assumptions to frame explicit theories, research and applications. New ontologies may improve problem solving within that domain. Translating research papers within every field is a problem made easier when experts from different countries maintain a controlled vocabulary of jargon between each of their languages.

Applications in genomics

• Gene Ontology: Genes and gene products in subdomains of BP, MF, CC -- biological process, molecular function, cellular component
• Human Phenotype Ontology
• UBERON - cross-species anatomy
• Cell ontology
• Cell line ontology
• EFO - experimental factor ontology

Tags from any of these can be encountered in various annotation resources.

rols: Basic idea

• OLS is ontology lookup service
• has API
• rols package help to interrogate the service
• ontologies are everywhere

Learn about 'smooth muscle' with rols

library(rols)
ss = OlsSearch("smooth muscle", rows=100)
ss
tt = olsSearch(ss)
dd = as(tt, "data.frame")
datatable(dd)

ontoProc -- capitalizing on ontologyIndex (thanks Daniel Greene!), Rgraphviz (thanks Kasper Hansen!)

library(ontoProc)
co = getOnto("cellOnto")
head(co$name)

The ctmarks app: walk through linked ontologies such as PR and present additional facets about the concept in focus

Limitation: the OBO representation in use is outdated and out-links are sparse

chk = ctmarks(co)

Projects: - use rols to get more interesting information about terms into the app - update the ontology resources - go beyond OBO ... but not all the way to OWL? Evaluate the UI/UX needed to broaden ontology usage - impacts: data integration, precision of annotation, cognitive efficiency

shinywow2 - check out vjcitn.shinyapps.io/tnt4dn8 but be patient and don't do it while i am doing it ...

vjcitn/CSHstats documentation built on July 31, 2023, 2:31 p.m.