Hi everyone,
Its been over 3 months that I am trying to develop a script for variant calling and RNA seq analysis for my project. I have attended quite a few workshops but it feels like a scam. I have nobody who can guide me and I really want to learn the analysis. Can anybody tell me if there are currently any short term courses for the same?
This is unfortunately the pain-point of bioinformatics journey which I've been through. I can recommend this book - https://www.biostarhandbook.com/ & have a look at the companion course https://www.biostarhandbook.com/edu/course/6/.
For variant calling (germline/hereditary) look into the GATK workflow https://gatk.broadinstitute.org/hc/en-us/articles/360036194592-Getting-started-with-GATK4
For RNAseq, I suggest you take a look at Salmon and Deseq-2 https://combine-lab.github.io/salmon/ & https://bioconductor.org/packages/release/bioc/html/DESeq2.html
For the Salmon and Deseq-2 combination this tutorial will help but you need to know some Linux & Rstudio https://www.hadriengourle.com/tutorials/rna/#import-read-counts-using-tximport
Most freeware stuff is optimized or available on Linux. Ubuntu is not a bad start. If you need a virtual machine, have a look into Docker or https://www.virtualbox.org/
You should look into develop basic skills in Linux, Rstudio and Python to run the tools. Nowadays, you don't really need java or C-level languages to run the tools. Developing algorithms are a different story though. Keep your determination, I knew nothing in 2017 when I started as well :)
Maybe some of the courses offered by the datacamp might help you to develop your bioinformatics skills. They are not for free but currently the offer is €149 for one year of unlimited use. see e.g.
https://www.datacamp.com/tracks/analyzing-genomic-data-in-r
I got to add some further comments with everything above in mind.
1) Try to write the code instead of copy and paste. Reason for this - develop muscle memory so it becomes second nature to you without thinking.
2) Try to learn the basics and not everything. There's no need to learn every single algorithm and coding languages. Focus on the workflows, algorithms and best practices in your field/papers from your peers. Most people use a fraction of what's out there and people have preferences of what kind of tools they use. That means you don't have to know every type of mapper, read trimmer, variant caller out there. Just know the relevant ones to your application.
3) Don't be afraid to experiment, keep practicing. You must practice constantly with coding so that what you learned doesn't disappear. You don't have to practice everyday, but for a few years you should consider a few practices every week until coding becomes second nature for the basics.
4) When you get an error and something doesn't work, look at it as a chance to learn and don't be discouraged.
5) If you don't have OSX/Macbook, this video tutorial by Networkchuck for creating a virtual machine will help you make a great Linux virtual machine to get you started. I don't recommend Windows for bioinformatics if possible.
https://youtu.be/wX75Z-4MEoM?si=sMlxiDnbyHQRvPCV
6) Bioinformatics is complicated field that intersects computer science, statististics & biology. As you have experienced it also requires good mentorship for someone to gain competence in. You are essentially learning two skills at once, 1) Learning to learn and 2) Bioinformatics. Give yourself time to take it all in, I'm 7 years in and still learning despite having worked in this area for 7 years.
Can Justin Kiessling Hi can, thank you so much for such a detailed answer. I would like to give some insights about what challenges I am facing. I tried learning it on my own but the drawback is that I get stuck in any error for weeks and months. The reason is that I am not from a computer I am background. Like the other day I could not understand why despite having python 3, its not taking commands for softwares that use py3.
I tried learning from different platforms and papers but the downside is that in each paper they have mentioned the pipeline specific to their understanding and at different platforms (workshops) they do not teach plant specific analysis, (mostly teach animal specific). Secondly workshops are so lame that the people teaching the code haven't themselves understood Linux properly. Everybody has their own methodology, there's nothing common such as the basics. Like some use galaxy some use jupyter notebook, but nobody teaches how to analyze data through basic Linux commands. currently I am stuck at the bwa mapping. I have a set of data but don't know how to proceed from there as I am confused after taking so many workshops.
I know what you mean, when I started in 2017 with my journey I encountered similar issues. It's a taste of what you will experience as a working bioinformatician, 1 day you might be doing RNAseq project, next day methylation sequencing which are two completely different workflows. And then someone will ask as a favor for a 16S/18S/ITS project to finish off their PhD thesis. What's the problem with bwa mapping?
Can Justin Kiessling after bwa mapping I need to check the QC of my mapping and the code that I am running is not well understood my me. after bwa mapping for different samples I will run a QC and then merge. I am stuck at the QC
How is the mapping rate, are you using Illumina or some other sequencer? What kind of QC were you thinking, where are you stuck in the QC?
Can Justin Kiessling sorry for the late reply. I got caught up. I have bwa sorted file. I have the code for qualimap, I just know how to install it, and run the code on Linux, also I have just copy pasted the code from somewhere don't know its authenticity at all. I can message you my script if you can check it once you have free time. Also, I want to learn how to annotate vcf files with data bases using GATK best practices for somatic mutations in plants.
I see, for annotating the somatic variants, what do you plan to use?
Can Justin Kiessling snpeff but I don't know the downstream process
Hi Can Justin Kiessling can you please share the biostar handbook e copy? if you have it can you share it?
It is a book that you will have to pay for to gain access. Honestly it is worth it given the content.
- Badges
- Science topic
- Similar topics
- Vectorization