Learn to use tools from the Bioconductor project to perform analysis of genomic data. This is the fifth course in the Genomic Big Data Specialization from Johns Hopkins University.
The Bioconductor Website
Loading...
来自 Johns Hopkins University 的课程
Bioconductor for Genomic Data Science
134 个评分
从本节课中
Week One
The class will cover how to install and use Bioconductor software. We will discuss common data structures, including ExpressionSets, SummarizedExperiment and GRanges used across several types of analyses.
与讲师见面
Kasper Daniel Hansen, PhDAssistant Professor, Biostatistics and Genetic Medicine
Bloomberg School of Public Health
In this lesson, we shall discuss the Bioconductor website.
[INAUDIBLE] I've navigated to the Bioconductor website.
And let's make it a little bigger so we can see what goes on.
The Bioconductor website contains a lot of useful information.
It's organized into these four different categories here with Install,
Learn, Use and Develop.
Under installation, there's a short
description of how to install Bioconductor that we will cover in a different section.
There's a lot of good information there.
Usually when I go to the Bioconductor website,
it's to find information about a given software package.
Down here under Use, we have software.
You can see the Bioconductor divides its packages into software packages,
which are the main workhorses.
Annotation packages with a pre-package annotation information from
various databases.
And finally, Experiment packages which contained data from populous tables,
which I used for tutorials, examples.
And sometimes to document an entire reproducible research work flow.
So let's click on the Software tab.
And here we get a list of all the Bioconductor packages in
the current release.
Out to the left here, we have a number of keywords.
These are called BioCViews or Bioconductor Views, and
they're keywords that package authors used to check the different packages.
The intention is, that this will allow you to easily find packages of interest for
a specific statistical method or biological question.
For example, if we expand the BiologicalQuestion,
we can see that there's 30 packages where the authors have
decided that they deal with alternative splicing and
there's 173 packages dealing with DifferentialExpression.
Let's find one of the packages and explore the package page a little bit better.
I'm going to go for the DESeq package.
That isn't under DifferentialExpression.
That's a little weird.
Let's see what other things we have down here.
Let's just expand out to the top and
over here in the search table, search for deseq.
So I click on the package, and this is the package homepage and
the release version of Bioconductor.
There's some standard information, a very brief overview of what the package
does and some authors and some citation information.
Down here under the package I see things such as which other
packages uses the package and what it's being used for.
But most importantly,
under the Documentation tab here, I can see a list of vignettes and NEWS files.
In this case, the package has a single vignette,
which is called Analyzing RNA-Seq data with the "DESeq" package.
Package vignettes is something that belongs in all our packages but
Bioconductor has mandated that software packages has a vignette.
And this is really the first place to go for information about a given package.
Some packages have very short vignettes,
some packages have extremely long vignettes, sometimes hundreds of pages.
And some packages have many different vignettes.
In this case it has a single vignette, and let's click on the PDF here, where we
get taken to a long description about what the package does and how to use it.
We won't really cover it right now, but
you can see here in the contents that the package has a lot of vignettes,
it's roughly 23 pages and has a lot of important information.
Let's scroll down here.
There's usually an example of an analysis on a specific data set,
there's some plots, and lot of really useful information.
This can be incredibly valuable.
I will always encourage any usage that start of by looking at the package
vignette when they're using a new package.
Another thing that fewer packages utilize but
which I sometimes find very useful, is the package NEWS file.
So the idea is that the package author can put in information about
what changes from version to version.
This can sometimes be a lifesaver and if there's a package you're using a lot and
that's critical to your workflow, it's not a bad idea to look at the NEWS
file every time a new version is released or a new version of Python is released.
Let's click on this.
This is a text file, and we can see here that the author has added
some information about what has happened in the various versions.
This is probably out of date.
You can see the last comment was added in 2012, second half of 2012.
I'm pretty sure that these package has changed since then.
You can also get package vignettes, they install into R.
So, if we for example switched over to R Studio for
a moment, here I'm back in R Studio.
Over here to the right,
I have the Help panel and I've gone to the Home part of the panel.
I can click down here on Packages.
And now I navigate the packages that I've installed in my local installation of R.
In this case here I find I don't have DESeq but let's take another one.
Let's take a package such as BSgenome.
Here's a index of all the functions in the package.
But up here at the top we have a link to a description file, we have the user guide
package, we have some other documentation, and the package NEWS.
So let's click on the vignettesm we can see this package has
two different vignettes, something called How to force BSgenome,
and then if it's in genome searching with and BSgenome.
Here we have it PDF, some source code and the R code.
There's also a link to the package NEWS.
Again, the author also started the NEWS file and
didn't really add a whole lot of information since he wrote it.
So these are two good examples I guess that,
the NEWS file can sometimes be of limited utility.
You can also get to this package,
[INAUDIBLE] list a little bit easier in our studio.
In our studio we have this paddle,
in this case in my setup I have it over to the right, that's called Packages.
I click on Packages and these are the packages that I've installed.
And if I click on it, say Bump Hunter, I get taken to the help page for
the package again and I can navigate down to the vignette.
Let's go back to the Bioconductor web page.
And let's get to the home of the page.
So, other useful information here, is down under the Learn side,
and I'm going to highlight this thing here called Common work flows.
Work flows is a new thing in Bioconductor.
The idea is that they detailed even bigger pieces of information that are packaged
vignette.
Work flows show how to use multiple packages together to achieve something.
Workflows is a new addition to Bioconductor, so we don't have
hundreds of workflows at the moment, but the ones we have are quite good, I think.
We are actually working on adding new workflows.
Let's click on one of them, and
here we have a list of Basic Workflows, Advanced Workflows.
Let's click on something like Sequence Analysis.
And we get a document, in this case, an HTML document.
And there's a fairly long description of how to use various packages together.
In this case here, this is a complicated diagram about which different
packages in R are used at different stages of a sequence analysis workflow.
And there's a lot of useful if short information down below here.
The idea is that you might go to a specific package in here to
find more information about a specific package.
This work flow that gives you some overlap, sorry, some over view.
We also in the Bioconductor host a lot of courses all over the world, short courses.
And here I click on the Course tab, and here's a list of all the different
tutorials, which we are allowed to post from recent courses.
A few months ago we had the Bioconductor 2015 conference and
here's a number of tutorials that was presented at this conference.
In this case they're either PDF or HTML.
These are written by experts in the field.
Let's click on the Flow Cytometry one.
And here there's a nice
little up-to-date work flow on how to analyze flows that turn in to data.
This is great for self-learning, but
the courses can be a little, it's not so well organized.
And it's most useful if you can find a recent course detailing
what you are interested in.
So here you have to do a little bit of searching and digging in
order to find the useful information, but there's a lot of good stuff here.
This is not the first place I'd go for information, but
it's there and it can be incredibly useful.
