- Видео 68
- Просмотров 2 248
Scientific Illustrations
Нидерланды
Добавлен 20 окт 2023
This channel is mostly about illustrations in genome biology using programming in R and Python. Based on the courses for PhD students in UMCG, University of Groningen
0.5. Make fastqc and multiqc reports in one python script
If you want to simplify the process of making fastqc and multiqc reports you can write a python script which will do it all. All what you need is present in the OS library of python.
#fastqc #multiqc #python #os
#fastqc #multiqc #python #os
Просмотров: 2
Видео
Happy New Year with python
Просмотров 614 часов назад
Just for fun. A small script in python which sends you the New Year greetings if the day is right. #python #happynewyear
0.4. Run multiQC with python
Просмотров 34День назад
MultiQC is a tool to make a single summary report out of multiple fastqc reports. Here I explain how you install multiqc in the controlled python environment. You can run it from any python IDE, but I prefer VSCode, in which you can write a line in. the command line or make a small script in python using function os.system() #multiqc #fastqc #python #vscode #fastq
0.3. Two ways of using fastqc to generate quality report of your fastq files
Просмотров 1014 дней назад
FastQC is a program that you can download and use for quality analysis of your FASTQ files. It can also be installed using Homebrew (brew) and run from the command line, allowing you to generate quality reports for multiple files simultaneously. #fastq #fastqc #homebrew
0.2. Download multiple files from SRA using shell script in macOS
Просмотров 1228 дней назад
In this video I explain how you make a shell script, when you need it and how you run it. As an example I show how you download multiple compressed sequence files from NCBI SRA, unzip it and convert to the final format fastq.gz with one push of the button on your computer. I wrote it in zsh shell language, which is currently default in macOS. For writing and executing I used Visual Studio Code,...
0.1. FASTQ from NCBI SRA
Просмотров 37Месяц назад
What if you want to get the very primary data from NCBI? You have to look for FASTQ files in SRA (Sequence Reads Archive). Instead of FASTQ you will find two alternative formats, normalised and lite, which can be converted to FASTQ (and gzipped) using sra-toolkit (with a function fastq-dump). Rather complicated process for seemingly simple task. This is an example for macOS computer. #FASTQ #sr...
1.11. CEL files in limma GEO2R script
Просмотров 48Месяц назад
What if you have raw .CEL files for expression microarrays. You can easily incorporate them into you limma script with justRMA() function. This and a few more tricks are in this video.
1.10. Multiple groups and contrasts in limma
Просмотров 212 месяца назад
You are probably used to compare too groups. But what if they are more than two? Don't be scared, it is actually fun and you can get more than you would expect. #r #limma #gene_expression #contrasts
3.3. Time in DESeq2
Просмотров 382 месяца назад
Can I use the time or other continuous variable in DESeq2 protocol like I did it in limma or edgeR/ The answer is "yes"and the trick is the same. You use either splines library or poly() function to transform your continuous variable into a factor, which can be used together with other categorical factors. #r #gene_expression #deseq2 #edger #time
6.1. Convert your Rmd script into a good looking report
Просмотров 643 месяца назад
When you want to share your script with others you can use *.Rmd file format, which is natural to R-studio notebook. After proper testing and running the lines you use KNITR function to make you nice report in three possible formats: html, pdf or doc.
3.2. Using time (age) scale in limma
Просмотров 443 месяца назад
Limma package is made for differential gene expression analysis with microarrays. Usually you do comparison of groups. Can you also incorporate continuous variables in your analysis? The answer is "yes". It's quite similar to options available in edgeR #r #gene_expression #limma #edger
3.1. edgeR with real age and gender. Will it work?
Просмотров 193 месяца назад
Can we combine continuous and categorical data in our model design and find differentially expressed genes? The answer is "yes". Although it will differ in numbers (compared to age groups) and the graphic output is more complicated. This video illustrates this option #r #gene_expression #edger #TMixClust #aging
3.0. Using time scale in edgeR differential expression protocol
Просмотров 564 месяца назад
Most of the differential expression analysis is performed in a context of a groups analysis. We compare male to female, old to young etc. But can we use for instance real ages and find out what is differentially expressed in time? The answer is "yes". One of the solutions is described in edgeR users guide. #r #gene_expression #differential_expression #edger #time
Install Hisat2 mapping program on your Mac
Просмотров 1734 месяца назад
Hisat2 is genome alignment program, the successor of TopHat and Hisat. It is not always clear how to install it on Mac computers, because most of the installations of this kind of the programs imply linux type OS. Here I used Bioconda recommended installation #genome_alignment #Hisat2 #TopHat #Bioconda #conda
2.3. EnhancedVolcano and other Volcano plots
Просмотров 534 месяца назад
I compare different volcano plots including EnhancedVolcano. I comment the name and the origin of the plot. Did you know that there are two different plots both maned volcano plot? R script is included #r #gene_expression #limma #volcano_plot #enhanced_volcano
2.0. Interactive illustrations in Glimma
Просмотров 1224 месяца назад
2.0. Interactive illustrations in Glimma
1.9. DESeq2 improved and with illustrations
Просмотров 275 месяцев назад
1.9. DESeq2 improved and with illustrations
1.6. Alternative glmFit based edgeR protocol
Просмотров 296 месяцев назад
1.6. Alternative glmFit based edgeR protocol
1.5. Classic edgeR protocol with one factor
Просмотров 36 месяцев назад
1.5. Classic edgeR protocol with one factor
1.4. Quick Start edgeR protocol with figures. Script based on the edgeR Quick Start lines
Просмотров 276 месяцев назад
1.4. Quick Start edgeR protocol with figures. Script based on the edgeR Quick Start lines
1.3.Quick Start edgeR. A protocol based on the Quick Start chapter in the edgeR users guide.
Просмотров 526 месяцев назад
1.3.Quick Start edgeR. A protocol based on the Quick Start chapter in the edgeR users guide.
1.1. limma-edgeR script. 2nd version. Major lines taken from limma tutorial
Просмотров 277 месяцев назад
1.1. limma-edgeR script. 2nd version. Major lines taken from limma tutorial
1.0. Introduction to RNAseq expression data and scripts
Просмотров 357 месяцев назад
1.0. Introduction to RNAseq expression data and scripts
0.8. voom or vooma? What is it and how to use it in gene expression analysis
Просмотров 327 месяцев назад
0.8. voom or vooma? What is it and how to use it in gene expression analysis
What is the key difference between glmfit and glmqlfit and how do you decide which one to do? Is it related to your design?
According to Gordon Smyth group publication from 2016 "the glmQLFit and glmQLFTest functions, which are alternatives to glmFit and glmLRT. They replace the chisquare approximation to the likelihood ratio statistic with a quasi-likelihood F-test, resulting in more conservative and rigorous type I error rate control." You can find this reference at the end of the edgeR user's guide. In short: glmQL version is the latest best option according to the authors. Is it really much different from the others I cannot say, it depends on the data
Thanks to you i managed to install STAR. Could you make a video about installing hisat ?
Thanks for the question. I just uploaded this video ruclips.net/video/YHD4jg7n9LU/видео.html
Thank you!
For nVennR, you should repeat the plotVenn command to get a more compact diagram (myNV <- plotVenn(nVennObj = myNV)
Thanks! It worked indeed
Thank you! I'll take a look at the sausage cloud.
Thank you Lenya!
Go Lenya! I am recommending this to all my students.
I would not say that non-parametric tests would work "better" in cases of non-normal distributions. Rather they will be always appropriate in these, albeit always giving you less statistical power compared to parametric tests. A sad example for many biologists is thag you will never get a statistically significant non-parametric test with < 8 data points per compared group. So, if someone can afford only triplicates there is literally no statistical way to prove significance, unless using something like T-test. However, in many cases (qPCR, blood meausures and etc.) data is normally distributed, saving these miserable data points in the eyes of reviewers... Anyway, totally agree, knowing your data type, its distribution in advance is the key for the research planning and implementation.
Thank you