It is amazingly well-produced and explained in detailed. I can see a logical order which makes perfect sense. I would like to see other methods, like ATAC-seq, DAP-seq, STAR-seq, etc. Please keep doing more!
First i was wondering why no one has mentioned the weird song at the beginning.but then i listened to the actual video and figured why...it was amazing...by far the best tutorial on this subject.
Wow. Josh, thank you for an incredibly helpful video. As a first year Ph.D student this eased/clarified concepts/procedures that I missed in the ChIP class. Thanks!
It's amazing how well you explain this, and not only this (I've seen other vids of yours). I'm a telecom engineer in process of turning into a would-be data scientist and I've understood it! BAM!! ;)
Thank you so much for making this so straightforward and easy to understand!!! It's so easy to find way too much info online that has loads of crap added in, but this allowed me to understand it nicely! Keep it up :D
Great video! I linked your video in my Medium article on predicting transcription factor-DNA binding using deep learning, and also my video (coming soon!) on the same topic in the description box. Thanks for the awesome explanation!
It would be lovely if you could explain the Model-based Analysis of ChIP-seq (MACS) algorithm for calling the peak. I'm sure it would be very clear explained by you. Please keep developing this channel, I've watched all of your videos and they are all great
@@statquest I know that your area of expertise (and the focus of this channel) is bioinformatics, however, videos explaining basic molecular biology techniques such as immunoblotting, RT-PCR etc. in your style of simplifying and clearly and concisely explaining the subject matter would be incredible...
I want to know how histones post translational modifications are identified or tagged? As methylation of histones will make the genome packaging tighter and acetylation did the opposite?
Would you potentially consider copy-number variant discovery using NGS data? I work at a university and most of our research staff no about copy-numbers and the data you get back, but have no clue how this data is derived. The resources available are pretty high level!
Hey, The lecture was great, i completely understood the concept of ChIP Seq, I have on doubt, lets say if the DNA binding protein is unknown, for example if its a novel transcription factor and we don't have much information about it. How can we raise antibodies against that protein if its completely new and also how can we identify the DNA sequence subsequently?
I could not follow why we have more reads mapped to a particular region of the genome when we wash away all the other parts but they don’t occur at the control experiment?
In the control, we don't have the antibody to detect DNA binding proteins. So we just sample random parts of the genome. In contrast, when we use an antibody for a protein that binds to specific parts of the genome, we will get reads stacked up in specific regions.
StatQuest with Josh Starmer Oh so we sequence same number of reads for both experiments and thus, the one concentrating on that specific regions has more reads on that regions? If so, isn’t it bias, since we will have orders of magnitude difference when it comes to the number of regions to be sequenced?
That's the whole idea. We measure the bias. In places where there bias is super extreme, we call that a "peak" and that suggests a location in the genome where the protein was bound.
Great video, as always :)! Helps me a lot how to phrase what am I doing.. but regarding the control, what would you like to recommend, GFP or total input in case of plants?
great video! I just have 1 question about the control track. why does it show very little reads, instead of lots of reads everywhere? for the actual experimental track, there’s a high concentration in that area because we’ve isolated DNA fragments that a particular protein binds to, and those DNA fragments occur near one another in the genome, right (is that what the x-axis, if we can call it that, of the track is? the location on the genome)? while in the control track, we didn’t isolate any DNA fragments. so why wouldn’t there be a large amount of reads everywhere on the control track, instead of a very little amount like there is in the video? thanks!
Both the treatment/experimental track and the control track have the same number of reads. If the experiment is done correctly, then for the experimental track, these should be concentrated in specific locations. For the control track, if everything went well, the reads should not be concentrated in specific locations, meaning they should be spread out throughout the entire genome. Since both tracks have the same number of reads, the ideal control track should be very low because there are so many places all over the genome to generate reads for. In contrast the experimental track has relatively few places in the genome to generate reads for, so all of its reads are concentrated and make tall peaks.
StatQuest with Josh Starmer ohhhhh that makes sense!!! I didn’t consider the fact that the number of reads are the same since we end up amplifying the wanted DNA in the experimental version anyway, so it goes from being super spread out (well, in the control track) to super concentrated. Thank you so much!!!!!!!!
Thanks so much for this important video. Now I have a clearer idea to ChIP-Seq. I am studying RNA-seq analysis and your profile is seen a "Holy Graal" to me, seriously. I have a request; I have a lot of problems understanding the main differences across "Sleuth" vs "DESeq2" and I believe that a stat quest would help us to have a full picture of the issue. THANKS Maria.
Thanks for your quick reply; actually this is already clear to me but since I am quite keen to understand some technical differences (even if I am not a bioinformatician) I tried to read this: www2.eecs.berkeley.edu/Pubs/TechRpts/2016/EECS-2016-131.pdf everything was quite understandable even for an idiot like me but actually going from 3.4 to 3.10 it was utterly hardcore...if you could help me that would be the best day since I have started my phd. Thanks
I have a question. How come that you have lots of DNA fragments after immunoprecipitation and crosslinking, when you already isolated the beads containing specific antibodies where your protein of interest is attached? Shouldn't you have only the DNA fragment where the protein is interacting?
Because 1) The protein of interest may bind many places in the genome and 2) The experiment is usually performed on millions of cells - so even if the protein binds a single place in the genome, we will still get millions of fragments (that all align to the same location).
@@statquest thanks for replying sir. so you're saying that all the DNA fragments that come out after reverse linking, would only be DNA fragments where the protein of interest attached to?
@@jimwelbryanchristopherferr3208 In theory that is correct, however, the process is not perfect and there is a little bit of contamination. That is why we then use "peak finding" programs like MACS2 to find enriched sites.
Hi Josh, I find your videos very helpful! thank you so much! I realized the the index doesn't contain this video. Would you mind updating the index some time?
![Felix "Unfair" | [Stray Kids : SKZ-PLAYER]](http://i.ytimg.com/vi/Oswujxm2Ag0/mqdefault.jpg)
Support StatQuest by buying my book The StatQuest Illustrated Guide to Machine Learning or a Study Guide or Merch!!! statquest.org/statquest-store/
You are a magical, magical human. Your videos are so incredible.
Wow, thank you!
Best explanation. After reading so many literature also, I am unable to understand fully. You explained in just 8 minute. Great job.
Thank you! :)
It is amazingly well-produced and explained in detailed. I can see a logical order which makes perfect sense. I would like to see other methods, like ATAC-seq, DAP-seq, STAR-seq, etc. Please keep doing more!
First i was wondering why no one has mentioned the weird song at the beginning.but then i listened to the actual video and figured why...it was amazing...by far the best tutorial on this subject.
Thank you! :)
Really living for the enthusiasm, Josh. Love the intro/outro.
Thanks! :)
Best explanation EVER. Thank you!
Hooray! :)
I feel like you explained this to us like we are five year olds which was really helpful. Thank you so much!
Glad it was helpful!
DUde, I watched ur videos to understand ML concepts for school, now I'm learning genomics from u. that's awesome thank u
BAM! :)
Wow. Josh, thank you for an incredibly helpful video.
As a first year Ph.D student this eased/clarified concepts/procedures that I missed in the ChIP class.
Thanks!
Glad it was helpful!
Hi Josh, can you do a statquest on ATAC-seq? Thank you very much!
same request!!
Or maybe even how CHIPseq and ATACseq can be used together?
same request still
Found this channel and instantly subscribed. The whole style of your explanations is great! Keep up the great work :)
Thank you very much! :)
Finest teacher for a reason.....
please do more and more videos on stat things in genomics.
love and respect from India.....
Thanks!
It's amazing how well you explain this, and not only this (I've seen other vids of yours). I'm a telecom engineer in process of turning into a would-be data scientist and I've understood it! BAM!! ;)
Hooray!!! I'm glad you like the videos! Double BAM! :)
Extremely comprehensive and lucid! Thanks!!!
Thank you very much! :)
it has never been so clear before... thank you!
Thank you!
Very beautiful and simple explanation. Thank you so much for that! It is going to help me a lot for my human genomics exam :)
Good luck and let me know how your exam goes. :)
Thank you so much for making this so straightforward and easy to understand!!! It's so easy to find way too much info online that has loads of crap added in, but this allowed me to understand it nicely! Keep it up :D
EXACTLY. And animations always help :D
that was the best explanation I´ve ever heard. You should teach some stuff to my professors!
Thank you so much! :)
best Intro from your Chanel so far
Bam! :)
StatQuest you did it again. You beautiful handsome human being.
Wow! Thank you very much! :)
You are the best, thanks to you I was able to understand all this in just few minutes.
Hooray! :)
thanks for making this simple and gentle, very effective!
You're welcome! :)
I LOVE THIS SIMPLE BUT INFORMATIVE PRESENTATION, I LOVE YOU JOSH!!!!!
Thank you so much!!! I'm glad the video was helpful. :)
Excellent and organized presentation. Thank you!
You're very welcome!
Best explanation ever...Loved it! Thank you so much for this.
Glad it was helpful!
i love the overly simple explanation with no soul or excitement, i finally understood, guess i am not the brightest lol
thanks a lot
noted
Great video! I linked your video in my Medium article on predicting transcription factor-DNA binding using deep learning, and also my video (coming soon!) on the same topic in the description box. Thanks for the awesome explanation!
you honestly saved me for my midterm
Hooray! This is great news! I hope you did really well.
It would be lovely if you could explain the Model-based Analysis of ChIP-seq (MACS) algorithm for calling the peak. I'm sure it would be very clear explained by you.
Please keep developing this channel, I've watched all of your videos and they are all great
I'll keep that in mind. :)
This was a HUGE help, thank you!
Hooray!!!!
This is incredibly well explained. Thank you!
You're welcome!
@@statquest I know that your area of expertise (and the focus of this channel) is bioinformatics, however, videos explaining basic molecular biology techniques such as immunoblotting, RT-PCR etc. in your style of simplifying and clearly and concisely explaining the subject matter would be incredible...
@@omairshariq6444 Thanks for the complement! If I have time, I'll take a stab at these molecular techniques.
Thanks a lot, Josh. Amazing as always :-)
Thank you! :)
Thanks so much for the explanation, it was really helpful!!
Thank you! :)
Awesome video! Helps a lot with studying!
Hooray!! :)
一開始唱歌那段就讓人心情大好，想繼續看下去：）
bam! :)
Really well done... Very gentle!
Thank you so much!!!!! You've save my day
Glad I could help!
the best, thanks for your work, really i like very much
Thanks a lot!
the introoooooooooooooo is LOVEEE
Thanks! :)
love how you explain!!
Thank you very much! :)
Thank you so much. This was great! Subscribed!
Awesome, thank you!
really made easy and informative;love it
Thank you! :)
Thank you so much! Your video is really helpful!
:)
legend! good luck with your music career xxxxx
Thank you!!! :)
just AWESOME. Thank you .
Thank you! :)
Thank you for the video! Very informative!
You're welcome! :)
Your videos are awasemone. Thanks
Thank you!!! :)
Thank you so much!!!! This is such a clear. explanation
Glad it was helpful!
Also if you ever want to do one on Cut&Run that'd be sweet! Thanks again for making these!
YOU ARE AWESOME! THANK YOU!
Thanks! :)
OMG I loved this so much
Thank you! :)
Thanks for this explanation
My pleasure!
Amazing. Thanks a lot!
Thanks!
Hi Josh, your videos are awesome! Could you consider doing one on ATAC-seq??
Sure! I'll put that on the to-do list, but it might be a while before I get to it.
Good！I understand how to read CHIP-SEQ！
BAM! :)
loved it...loved it... loved it
Hooray! :)
I want to know how histones post translational modifications are identified or tagged? As methylation of histones will make the genome packaging tighter and acetylation did the opposite?
There are antibodies to specific post translational modifications - so you can use this method to target those modifications.
haha the starting song was so interesting
:)
no one teaches better than you!
Thanks! :)
@@statquest 😊♥️👏
Would you potentially consider copy-number variant discovery using NGS data? I work at a university and most of our research staff no about copy-numbers and the data you get back, but have no clue how this data is derived. The resources available are pretty high level!
I'll keep that in mind.
Hey, The lecture was great, i completely understood the concept of ChIP Seq, I have on doubt, lets say if the DNA binding protein is unknown, for example if its a novel transcription factor and we don't have much information about it. How can we raise antibodies against that protein if its completely new and also how can we identify the DNA sequence subsequently?
There may be methods that can just determine protein-bound regions, in a general sense.
I really liked it but at 7:13, it not a kidney picture but a stomach 😂
True! Oh well. I need an editor ;)
You are really funny man, thank you Josh! BAMM
I could not follow why we have more reads mapped to a particular region of the genome when we wash away all the other parts but they don’t occur at the control experiment?
In the control, we don't have the antibody to detect DNA binding proteins. So we just sample random parts of the genome. In contrast, when we use an antibody for a protein that binds to specific parts of the genome, we will get reads stacked up in specific regions.
StatQuest with Josh Starmer Oh so we sequence same number of reads for both experiments and thus, the one concentrating on that specific regions has more reads on that regions? If so, isn’t it bias, since we will have orders of magnitude difference when it comes to the number of regions to be sequenced?
That's the whole idea. We measure the bias. In places where there bias is super extreme, we call that a "peak" and that suggests a location in the genome where the protein was bound.
3:15 why need to cut to 300bp fragments?
Because the machine can only handle relatively short fragments.
ATAC-seq would be awesome!
I'll keep that in mind.
wonderful, thank you friend - hi from Iraq .............................. jkjk southern USA
You're welcome!! I hope hurricane Frances didn't give you too much trouble.
Great video, as always :)! Helps me a lot how to phrase what am I doing.. but regarding the control, what would you like to recommend, GFP or total input in case of plants?
great video! I just have 1 question about the control track. why does it show very little reads, instead of lots of reads everywhere? for the actual experimental track, there’s a high concentration in that area because we’ve isolated DNA fragments that a particular protein binds to, and those DNA fragments occur near one another in the genome, right (is that what the x-axis, if we can call it that, of the track is? the location on the genome)? while in the control track, we didn’t isolate any DNA fragments. so why wouldn’t there be a large amount of reads everywhere on the control track, instead of a very little amount like there is in the video? thanks!
Both the treatment/experimental track and the control track have the same number of reads. If the experiment is done correctly, then for the experimental track, these should be concentrated in specific locations. For the control track, if everything went well, the reads should not be concentrated in specific locations, meaning they should be spread out throughout the entire genome. Since both tracks have the same number of reads, the ideal control track should be very low because there are so many places all over the genome to generate reads for. In contrast the experimental track has relatively few places in the genome to generate reads for, so all of its reads are concentrated and make tall peaks.
StatQuest with Josh Starmer ohhhhh that makes sense!!! I didn’t consider the fact that the number of reads are the same since we end up amplifying the wanted DNA in the experimental version anyway, so it goes from being super spread out (well, in the control track) to super concentrated. Thank you so much!!!!!!!!
That was stomach, not kidney at 7:13
Thanks so much for this important video. Now I have a clearer idea to ChIP-Seq. I am studying RNA-seq analysis and your profile is seen a "Holy Graal" to me, seriously. I have a request; I have a lot of problems understanding the main differences across "Sleuth" vs "DESeq2" and I believe that a stat quest would help us to have a full picture of the issue.
THANKS
Maria.
Thanks for your quick reply; actually this is already clear to me but since I am quite keen to understand some technical differences (even if I am not a bioinformatician) I tried to read this: www2.eecs.berkeley.edu/Pubs/TechRpts/2016/EECS-2016-131.pdf
everything was quite understandable even for an idiot like me but actually going from 3.4 to 3.10 it was utterly hardcore...if you could help me that would be the best day since I have started my phd.
Thanks
That's sounds great Dr Starmer. I will dive into these videos that I had had the chance to see yet! thanks!!
well explained mr english man
Thanks
perfect narrative...
hi, is there a PDF file that I can download? I would like to print some images to tape into my notebook
I have a question. How come that you have lots of DNA fragments after immunoprecipitation and crosslinking, when you already isolated the beads containing specific antibodies where your protein of interest is attached? Shouldn't you have only the DNA fragment where the protein is interacting?
Because 1) The protein of interest may bind many places in the genome and 2) The experiment is usually performed on millions of cells - so even if the protein binds a single place in the genome, we will still get millions of fragments (that all align to the same location).
@@statquest thanks for replying sir. so you're saying that all the DNA fragments that come out after reverse linking, would only be DNA fragments where the protein of interest attached to?
@@jimwelbryanchristopherferr3208 In theory that is correct, however, the process is not perfect and there is a little bit of contamination. That is why we then use "peak finding" programs like MACS2 to find enriched sites.
@@statquest I see I see. Thank you sir!
Hi Josh, I find your videos very helpful! thank you so much! I realized the the index doesn't contain this video. Would you mind updating the index some time?
Thanks for catching that mistake. I've updated the index. :)
Amazing video as always, but it's Lung vs Stomach at 7:13 lol
Yep! :)
you're great. thank you
Thank you so much! :)
Can you please make a video on NET-Seq
I'll keep that in mind.
Thank you!
You're welcome!
You are amazing! Thank you!
Thanks! :)
Clear cut picture
:)
Thanks a lot !
:)
really cute story
:)
just perfect
Hooray! I'm glad you like this video. :)
thank you a lot
You're welcome!
You rock!
Thanks!
Best!!!
Thanks!
Thank you my god
:)
lung vs stomach，Joshua
you are awesome
Thank you! :)
2:57
Thanks man...
Happy to help
thanks
You're welcome! :)
BAM!!!
:)
uf gracias loko, andaba jodido
No problema!
awsome
Thank you! :)
WOW
:)
wow
:)
i fucking love you