I really appreciate your kind words. Thank you very much :)

@@Bioinformagician I like your way of explaining 😻

I am also a newly enrolled student from India doing PhD in Bioinformatics its been only two weeks since i joined its very difficult at the moment since my background has always been of purely biotechnology but maam's videos has helped to clear many of my concepts till now and she teaches very nicely ! learning through her videos is a great relief and I hope you are in a much better position now !

I am glad to know this was informative ❤️

Thank you, I am glad you find my videos helpful! 🤗

I'm so glad!

Good idea, I will work on creating such a playlist. Thanks for the suggestion!

Hi, I am glad you found my video helpful! I am planning to create more videos to perform various analyses on RNA-Seq data. I uploaded a video yesterday that shows how to perform data manipulation on the same data in R using dplyr (kzfaq.info/get/bejne/aqmbhbt4vNOlkok.html), if that is of interest, make sure you take a look!

​@@Bioinformagician hi thank you for this informative videos. Can we extract the expression data for few genes using GEO database.

Yes, you can perform meta-analysis using various RNA-Seq datasets. Check out this package in R to perform such analysis - www.metafor-project.org/doku.php/metafor

@@Bioinformagician madam, can't we do with deseq2

I appreciate your kind words. In order to deal with count tables, one can use a functional programming language like R, Python or Perl that provides data structures like datatables/dataframes which will provide the right functions and methods to manipulate (like filter columns) your data.

Thank you for the suggestion, will surely think about making a video on this soon :)

I am also interested in learning how to search wisely. I am looking for data sets to pull down for my students to analyze.

Usually, my selection for datasets depends on the associated publication. If you are looking for some breast cancer datasets, it might be helpful to look for papers that have conducted comparative studies using breast cancer data. In addition, I also prefer to look for papers published in high-impact factor journals, these papers have gone through a rigorous review process, which adds more confidence in the data and findings associated with the paper. Another measure would be to look at the number of times the associated publication has been cited. The next consideration while choosing a dataset is the expression unit, as you mentioned more often it so happens that datasets are in different units. In such a case, either you can use counts and normalize them to the same units or get the raw data (RNA-Seq reads) and process data using standardized pipelines and quantify gene expression to the same units, I prefer the latter. Lastly, there are large public data repositories like the TCGA data portal (portal.gdc.cancer.gov/) which hosts a lot of data available for anyone to access. You will find a lot of breast cancer datasets (in addition to many other histologies) and data is available in various units like counts, TPM, FPKM/RPKM. A plus of using the TCGA portal is it collects raw data from many cancer projects and processes them using standardized pipelines which makes comparison across different sources/projects sensible and valid. Hope this helps! :)

@@Bioinformagician wah! thank you so much for this!! your entire channel is a lifesaver to my undergrad thesis!!! will surely add you to my dedication :

@@Bioinformagician And just to share, I was planning to use series from GEO as my study/training cohort, and TCGA dataset for my validation cohort, but I am honestly still confused on how a validation cohort works, but I will hopefully figure it out along the way!

Use R

I surely plan on creating a video on RNA-Seq pipeline, from alignment to gene quantification step. Since I have a mac, unfortunately I cannot demonstrate workflows in a windows system. However, if you encounter errors, do post them. There are a lot of windows users, someone will be able to help you.

Most videos do not follow any predetermined order. The ones that are intended to be followed one after the other has been named as PART 1 or PART 2.

Thanks for the suggestion. I'll surely think about making a video covering this topic.

It's in my pipeline. Will soon be making a video covering this topic. Thanks!

Noted! Thank you!

Your criteria to select a dataset can depend on so many things like the experimental design, animal model, disease studied..etc. Regarding the RNA-Seq data repositories, there are other sources as well like Expression Atlas from EBI (www.ebi.ac.uk/gxa/home/) where you can get RNA-Seq data. There are also various publications where RNA-Seq data is generated and studied.

If no information is provided about the kind of normalization performed, you could either reach out to the authors who have submitted this data or you could see if they have provided raw data in supplementary files. If they have, you can process and normalize it for your downstream analysis.

@@Bioinformagician Do you plan on making such tutorial on how to normalize data in R..

@@syedmansoorjan2671 Yes, I plan to talk about different methods of normalization in a future video.

@@Bioinformagician That's cool, I expect as you did in previous videos, from basic to advanced..

linuxize.com/post/how-to-extract-unzip-tar-gz-file/#:~:text=Conclusion,followed%20by%20the%20archive%20name.

You cannot combine datasets from microarray and RNA-Seq and normalize them and make comparisons between them as they are completely different technologies. Having said that, you can process them separately, compare the results and/or perform meta-analysis on the results.

I looked up this record on NCBI GEO and it seems the supplementary file (ending in _RAW.tar) contains data about signal intensities from scanned images processed using Feature Extraction software. I am assuming you are interested in getting the normalized and log2-transformed expression data for all samples. You can get that by executing the following commands in RStudio: library(GEOquery) geo_id

Very informative video. Can you make some video on metatranscriptomics

@@tulikabhardwaj484 I will surely think about making a video on it soon :)

​@@Bioinformagician thanks a lot. it's a helpful tutorial. However the object, data, obtained by this method, i.e. data

@@yeunglee2565 I am not sure if I understand, can you please explain what do you mean by different? Do you mean you don't get the expression data when you run this command?

Thank you for the suggestion. I will ensure to put numbering in my video titles when they are in sequencing order. Most videos on my channel do not follow any pre-determined order.

You could refer to the associated publication and read methods section to see if they have mentioned what normalization was performed. If not, you might have to contact the authors for that information.

@@Bioinformagician ok, thank you, Maam.

Can you share the exact error you see when you try to install limma?

If you observe batch effects in your data, then yes, you should correct for them.

There are a lot of open source repositories which make de-identified data from patients accessible to the public. Check out - portal.gdc.cancer.gov/, xenabrowser.net/datapages/, www.cbioportal.org/ and lastly the publications that generate data serve as the best source for data.

It totally depends on what data you will be dealing with, what analysis you plan to perform, and the tools you intend to use. Your current configuration is good enough to be dealing with smaller data.

​@@Bioinformagicianvery Thank you, useful comment Do you recommand using x3dna?

@@pc66ghz I have close to no experience in structural bioinformatics. Unfortunately I cannot provide any recommendations for those tools. Sorry!

There will be a steep learning curve. With perseverance and decent googling skills, it is doable.

.CEL files are raw files from microarray data. You will need to perform an RMA normalization on raw data present in .CEL files. Here's a nice step-by-step tutorial: felixfan.github.io/RMA-Normalization-Microarray/

@@Bioinformagician thank you so much,

Unfortunately I do not provide tutoring. I intend to share my learnings via my videos. Happy to answer any questions you may have :)

Happy to help to the best of my knowledge! You can find my email in the description section of every video :)