Bioinformatics - KEGG Pathway Visualization in R

[@abdelrahmanmahany133]

It is a really great tutorial and I leaned a lot from you Alex. I will keep going deep in this playlist and waiting for new videos and tutorials.

[@goktugozkurt4281]

Thank you for your time. It was very helpful for me to understand the pipeline of RNA-seq. Especially, KEGG pathway analysis part was very helpful. Best wishes.

[@eigenmanncaroline2427]

Super video and easy to understand !! Got really cool results :)

[@HeLiAn5]

Thank you so much for all the lectures! they are really helpful!

[@lesterpascua5284]

I've just received transcriptome data for KEGG analysis and I have no idea how to do it. Searched YT and this is the first thing I see! Uploaded just a few hours ago. Thanks for this resource!

[@alexsoupir]

Haha good to hear! Hopefully they are helpful. Depending on what the source organism is it may be just as easy or much more difficult. For the gageData it has human, mouse, rat, and yeast based on the manual as well as orthologs. By reading a little bit more you should be able to pull data for others, those 4 are just the easiest to use since they are right there.

[@lesterpascua5284]

@@alexsoupir thanks for the tip! Appreciate it. Am working on plant conifer species at the moment. :D

Dear Alex, I have watched complete series. hats off to ur efforts and vision. Its the best, awesome, perfect and professional. please keep it high as this pipeline we have enjoyed. I expect that ur channel will grow and millions of people will get benefit from it. best wishes.

[@dvdmarotta]

Hi Alex, thank you so much for the amazing video-very helpful and clear! I have got my GO up and Down and I would like to create a table with those (separately) and also I would like to plot them in a graph. any hint ho how to do it? Thanks!

[@tinacole1450]

I am looking around for GSEA gene enrichment analysis. Gage looks like a great alternative.

[@Muuip]

Great presentation! Much appreciated!👍

[@iot3136]

Hi Alex, hello again. Could you please do me a favor. How can I get a KEGG Pathways for the whole genome of Arabidopsis thaliana (TAIR Genes). I want to write a local csv file combining both KEGG PW and TAIR genes for future use. Do I have to have a subscription for KEGG

[@alexsoupir]

Your comments show up in my inbox, but then I can't see them.. I'm glad they eventually show though! Since arabidopsis is super common and a model organism, i would be surprised if there isn't a way. There is an organism data base called org.At.tair.db, so as long as you're using or are able to map whatever your gene ids are to the tair identifier, you might be able to pull them? There's a manual for org.At.tair.db too that you could look into and see if there are special ways of doing the pathways or if it's the exact same as with human or mouse.

[@freezingtolerance7493]

Hi. Alex. Thank you for your video always. Since I am using "arabidopsis data base" which is org.At.tair.db... function(pid) is not working wherein the code set for annotpkg is "org.Hs.eg.db". How could I revise this code?

[@AA-gl1dr]

Thank you so much!

[@learningtime1367]

What is the difference between GO analysis and gobpres (line 331 of code)?

[@learningtime1367]

While doing the pathway analysis you include the whole differentially expressed genes and their log2foldchanges (line 341), why don't we just put up-regulated or down-regulated gene sets to see the pathways they are involved in and why do we do $greater to get up-regulated?

[@genticswithkazan]

Thank you again! in the data upload step. i am working on plasmodium falciparum. How can I find the that data?

[@vpeska]

Hi Alex, thank you so much for the great vids! You made my scope much broader. Could you give me a hint how to get go.sets.xy and go.subs.xy for bee (Apis mellifera)? I got stuck in visualization because I do not know how to make or pull them.

[@alexsoupir]

Hey! That's actually a great idea for a new video (been having a hard time thinking of a broad applicable video instead of ones that would help few people). There is a way to pull it in using BioMart that I can't remember exactly off the top of my head. Unfortunately, it isn't available for every host, but there a fairly large number. This is similar to the org.Mm.eg.db. maybe through some digging you can figure it out if i don't get around to making a video but I'll try.

[@vpeska]

Thank you @@alexsoupir I am trying my best, I used AnnotationHub to get org.Apis_mellifera.eg.sqlite, which is an OrgDb object usable for extracting GOs of up/down regulated genes. Nevertheless, I still cannot find an easy way how to prepare the sets/subs.

[@iuliagrigore4721]

Hey Alex! Amazing videos, thank you for posting them! For the majority of the keggres$greater or $less, I got the q.val approximately 0.8-1.0. What do you think went wrong? I only changed the '.mm' into '.hs' as I am using human data. In rest, I have copied everything that you have been doing in the tutorial. Best

[@Gbemi78]

Thanks for the upload. Please how can one draw the Kegg enrichment pathway and obtain the graph? I did mine but was giving my q value a value of 1, so am a bit confused. I am working on wheat and I was able to get the GO analysis but the KEGG pathway enrichment analysis is not just working out. someone recommended a package of keggrest but I cant still figure it out

[@alexsoupir]

Hmm. I haven't ever worked with wheat so I can't say for certain, but with the popularity of the organism I would be surprised if there isn't something in the KEGG grouping for it. Also, are you referring to like the FDR q-value being 1? That would mean that there aren't any significant genes after adjusting the p-value for false positives. If all of your genes have the adjusted p-value of 1, I think there might be something else of issue in the analysis.

[@Gbemi78]

@@alexsoupir thanks for your reply. First I used the DSeq r package to calculate the logfoldchange, p-value and adjusted p-value and I had genes that were both down and upregulated. Then I used the amino acid sequence of these genes to get the ko number which I ran on an online software at www.omicshare.com/ and gave me the result that the q value was 1. I don't know if I can get your email so I can show you the result of my analysis and the graph result. Anticipating your response. Thanks once again

[@justindhindsa4452]

Hi Alex! Thanks so much for your incredibly helpful videos. When you do this pathway analysis, do you typically use a Log 2 fold change threshold?

[@alexsoupir]

Howdy! When looking in the past I just use the significance as a cutoff, but if you want to be more stringent you sure can add an LFC threshold, too. However, doing this thresholding you may lose the gradient of significant genes that are closer to zero (may jump from white or zero to either bright blue or red, depending on threshold set and the extremes of your LFC values). If your LFC max is say 3, but you set your LFC threshold at 1, you'll "lose" 33% of the gradient from 0 to +/- 1 if I'm not mistaken, even if the genes do have a significant p-value? I could be wrong here though.

[@justindhindsa4452]

@@alexsoupir Awesome thanks! One other question for you: do usually produce bar graphs for the gene ontology terms/Kegg pathways (like you can in EnrichR)? If so, is there a specific package you use for that?

[@sanjaisrao484]

Sir its coming pathway may not exists Wht to do

[@iot3136]

Hi Alex, New Subscriber here :) Could you please create a short lesson on R for Bioconductor summarizedexperiments? Especially explaining S4 SE data analysis. Would be really appreciated. It looks your KZbin Channel has great stuff where no other KZbinr has done.

[@alexsoupir]

Hey! I saw the notification for your comment but couldn't find it. So glad I kept looking! I can look into summerizedexperiments and see what I can figure out! I've personally never used it but the goal of this channel was to help people who seemed to be asking the same question but couldn't find a resource to answer it. Hopefully able to fill the gap for many who cannot find the answer anywhere else! :)