The Three Tables in Genomics (2:10)

Loading...

來自 Johns Hopkins University 的課程

Statistics for Genomic Data Science

129 個評分

Johns Hopkins University

Statistics for Genomic Data Science

129 個評分

課程 7（共 8 門，Specialization 基因组数据科学）

An introduction to the statistics behind the most popular genomic data science projects. This is the sixth course in the Genomic Big Data Science Specialization from Johns Hopkins University.

從本節課中

Module 1

This course is structured to hit the key conceptual ideas of normalization, exploratory analysis, linear modeling, testing, and multiple testing that arise over and over in genomic studies.

Module 1 Overview (1:07)1:07

Reproducible Research (3:42)3:42

Achieving Reproducible Research (5:02)5:02

R Markdown (6:26)6:26

The Three Tables in Genomics (2:10)2:10

The Three Tables in Genomics (in R) (3:46)3:46

Experimental Design: Variability, Replication, and Power (14:17)14:17

Experimental Design: Confounding and Randomization (9:26)9:26

Exploratory Analysis (9:21)9:21

Exploratory Analysis in R Part I (7:22)7:22

Exploratory Analysis in R Part II (10:07)10:07

Exploratory Analysis in R Part III (7:26)7:26

Data Transforms (7:31)7:31

Clustering (8:43)8:43

Clustering in R (9:09)9:09

與講師見面

Jeff Leek, PhD
Associate Professor, Biostatistics
Bloomberg School of Public Health

The processed data from a genomics experiment can usually be stored in

three tidy data sets.

The first tidy data set is information about the phenotypes or about the samples.

So that's this data set up here.

So that data set usually contains information about,

say, who are the cases and who are the controls.

Information about, say, the batches that the samples were collected in.

Or any other technical or

biological information that will be used in modeling the genomics information.

The second data set is the genomics data.

So the genomics data is usually stored in a large data set like this.

It's usually tall and skinny.

In other words, there are usually a relatively few number of

samples compared to the very many number of features that have been collected.

And here by features I mean genes whose gene expression we've measured or

SNPs that we've measured and crossed with a large number of individuals or

any other information that's genomic that's been collected.

Finally, the data set, the features data set tells you a little bit

about the features that we've collected.

So, it tells you about which genes you've measured or which SNPs you've measured in

the population, and maybe some information about what genome they came from or

any other information you might want to know about those.

Say, what biological pathways they belong to or anything else.

So an important thing to know is that a first check you always want to do

is make sure that there are the same number of rows in the phenotype data

as there are columns in the genomics data.

Because the phenotype data represents the same samples

that appear in the genomics data.

Second, you want to make sure that there are the same number of rows in

the features data as there are rows in the genomics data, because the features data

says something about each of the features in the genomics sample.

If those dimensions don't match, you know that you've got some kind of problem

with some of the annotation in one or the other.

Now, there should always be sample indicators.

So, these are sample names that appear in this data set

that match the sample names that appear in the genomics data.

Similarly, you should have gene or SNP, or whatever other genomic feature IDs here

that should match the gene and SNP and feature IDs that appear here.

These are, again, basic checks that you can do to make sure that the data set has

been processed and given to you in a way that you can handle.

探索我們的目錄

免費加入並獲得個性化推薦、更新和優惠。

Coursera 致力於普及全世界最好的教育，它與全球一流大學和機構合作提供在線課程。

© 2018 Coursera Inc. 保留所有權利。

通過 App Store 下載

通過 Google Play 獲取