Join Dr Alon Goren as he discusses how to analyze your ChIP data and carry out genome-wide mapping.
Alon Goren completed his Bachelor's, Master's and PhD in Medical Sciences at the Hebrew University of Jerusalem, Israel. His PhD dissertation was focused on the role of epigenetic mechanisms during mammalian development.
Alon is currently a joint postdoctoral Charles H. Hood fellow at the Brad Bernstein and Aviv Regev groups affiliated with Harvard Medical School, Massachusetts General Hospital and the Broad Institute of Harvard and MIT.His current research combines experimental and computational approaches, and focuses on understanding genome-wide chromatin regulation during mouse in vivo early development.
Ladies and gentlemen, thank you for standing by and welcome to today's presentation: Exploring genome-wide Organisation of Chromatin Structure by ChIP. Your host today, Sarah Dolny, Marketing and Event Coordinator at Abcam. I would now like to turn the conference over to Ms Dolny.
SD: Hello and thank you for joining us for today's webinar: Exploring genome-wide Organisation of Chromatin Structure by ChIP. It is my pleasure to introduce today's presenter, Alon Goren. Alon acquired his Bachelor's, Master's and PhD in Medical Sciences from the Hebrew University of Jerusalem, Israel. His PhD dissertation focused on the role of epigenetic mechanisms during mammalian development. He is currently a joint postdoctoral Charles H. Hood fellow at the Brad Bernstein and Aviv Regev groups affiliated with Harvard Medical School, Massachusetts General Hospital and the Broad Institute of Harvard and MIT. His current research combines experimental and computational approaches, and focuses on understanding genome-wide chromatin regulation during mouse in vivo early development. Joining Alon today is Miriam Ferrer; Miriam completed her PhD at the Free University in Amsterdam working on cancer gene therapy, and did her postdoc at the MRC Laboratory of Molecular Biology in Cambridge where she studied the role of BRCA1 in DNA repair. She joined Abcam in 2008 and has been a product manager for assay kits since July 2011.
As Nathan highlighted earlier, if you have any questions throughout this presentation we invite you to submit them on the right hand side of your screen in the Q&A panel. Questions will be answered during the troubleshooting portion of this webinar. At this time, I'd like to hand the presentation over to Alon.
AG: Hello, and good afternoon to everybody. Thank you Sarah for the introduction, and Abcam for the opportunity to present here today. I also wanted to thank you all for joining us and listening to this webinar. First, I want to give you some kind of an overview of the talk today and the way that we're going to go through it. I want to first give an introduction to chromatin, to ChIP, which is Chromatin IP and different uses of the method. Then I will go over genomic approaches to map in vivo chromatin structure. Furthermore, I will go into a detailed description of the chromatin of the genome-wide chromatin organisation, and, importantly, I will describe recent and good usage of visualisation tools. I will then go and describe to you some major scientific discoveries that came from mapping of in vivo chromatin organisation, and then I will relay the talk to Miriam from Abcam. Finally, I will come back again to answer any of your questions. I will try to pause after each of the sessions to allow you to also have some time to post some of your questions, if you have any. Thank you.
Part I: I want to introduce you to chromatin and the way chromatin is organised in the genome. Recently, there was a really nice publication by a group which is called ENCODE, and I will revisit the idea for ENCODE later on, but I want to take this good set, good figure, good illustration in order to give you an idea of how chromatin is organised, and how can we learn about the organisation of chromatin? On the right side of the figure you can see the chromosome and as it slowly becomes opened up, you see that there are long-range chromatin interactions. Then you can look at the DNA which is wrapped around these balls, which we call nucleosomes. These bar-like nucleosomes are composed of histones, and when you consider the nucleosomes and the histones, there are a variety of mechanisms for their marking and their organisation that allows the organism and the cell to regulate gene expression via chromatin organisation. For instance, there is compaction and accessibility; compaction or accessibility of genomic regions, allowing for transcription machinery and transcription factors to bind, and this can be visualised by DNA as one sensitivity.
There are also a lot of modifications to the histones that can be learnt about by ChIP, as I will talk about in a bit. There's also DNA methylation which our modification to one of the nucleotides composing the DNA itself, and also involved in the regulation of gene expression. Another good way to look on a study, the organisation of chromatin, is to consider it as superimposed layers. On the left side we have the 'OFF' genes that are repressed and not transcribed; while on the right side what we put was the genes that are transcribed, or 'ON'. So if we consider first the DNA, it can be either methylated, as I mentioned earlier, or non-methylated. Usually, methylation goes along with repression of gene expression, then the next level is the nucleosomes and they vary both in terms of the composition of the histones that make them. Also you have some variations within the histones themselves, and the way that they're organised on the genome. Then if we consider the next level, is what kind of modifications the tails of the histones that put together and make the nucleosomes; what kind of modifications these tails have?
For instance, on the right side there are modifications that are mainly associated with active genes. For instance, H3K4me3 is usually found on active promoters, while H3K9me3 on the far left side is usually located on very repressive heterochromatin regions. The next level is a compaction where people - even before they have the ability to look on the level of the DNA sequence itself, and use microscopy - have been able to visualise that the chromatin is composed of hetero- and euchromatin. Heterochromatin is usually the more compact, while the euchromatin is more open and, for instance, one of the things that is depicted in this figure is that it's more open and allows transcription factors, or RNA polymerase II to bind in an easier way. Finally, one of the new things that has been recently learnt is that there is an order of the chromatin within the nucleus itself, what we call high order structures. For instance, in the repressive on the left side we can see that the chromatin can be tied to the nuclear lamina. On the right side, people have identified transcription factories which are an area of the nucleus that are composed from a variety, from a high number of transcription machinery such as RNA polymerases sitting, and the DNA areas that are transcribed are getting into these areas.
After I've introduced to you the way the chromatin is organised, I want to tell you a bit about the methods for learning about it. A very good method, and I think if I generally say it's one of the most common in vivo methods to study the association of proteins with DNA, is called ChIP. ChIP stands for Chromatin Immunoprecipitation, and in recent years people have been using it to map the organisations of system modifications, of transcription factors and of chromatin regulators. I will try to visit each of these types of proteins and the way that we have learned about them later on, but I want, again, to tell you a bit about ChIP first. ChIP was initially developed as early as 1988 by the Solomon and Varshavsky group, and they followed even earlier ideas that were set in 1984 by Lis and Gilmour. The basic idea that follows, if we take the cells and we cross-link the DNA to the proteins that are in association with it, and then we break this chromatin which is a combination of DNA and proteins to small enough fragments, we can then pull them down using antibodies.
This allows us to fractionate the genome into areas that are in association with the protein of interest, and areas that are not. Then the next step is to take the DNA that was in association with the protein of interest, and then look into which genomic areas are there? One of the very exciting ways to do that in recent years is to look at it on a genomic level using microarrays or sequencing, and this way to generate maps that provide the distribution of proteins throughout the genome. I want to cement the idea of ChIP and I will use another way to present it, and I also want to bring in a bit of the discrimination between ChIP-chip and ChIP-sequencing. In general, if we here on the top consider the DNA in black wrapped around nucleosomes that have tails that can be modified, here the modifications are either green or blue. Let's say that we want to figure out which genomic parts, which loci are in association with nucleosomes that have a modification, a green modification on them.
What we do is we cross-link the DNA to the proteins and this allows us to take a snapshot of the situation in vivo, and then we cut it into small fragments and this allows us to take an antibody that recognises specifically the protein of interest, or in this case the modified nucleosome tail, or histone tail that we're interested in. Then we pull down the antibody and the antibody is now connected to the modified tail, which is in itself connected to the DNA. This way we're going to pull down only the genomic regions that were in association with this nucleosome that had the modified tail. Then the next step is to either do the ChIP-seq or the ChIP-chip, and this, in general, has two different directions. First, you need to purify the DNA and on this end it's very similar if you do ChIP-chip or ChIP-sequencing. For ChIP-chip, which stands for using an array, which is the second name is ChIP, what you do is you amplify the DNA and then you label it with a way that you can then detect by fluorescent. Then according to hybridisation to array of probes, you can say if a specific genomic region was enriched or depleted in your sample.
This is a very good way to learn about the genomic organisation, or at least subgenomic organisation of chromatin, and it was used for several years in a very robust way and taught us a lot about the organisation. In approximately 2007, there was introduction of ChIP-sequencing and here the approach is it very nicely takes into account that you don't need to previously define which genomic regions you're interested in, you just take the DNA and sequence it. So there are several approaches to generate a lot of sequences, and this we call high-throughput sequencing, or next-generation sequencing. This way you can take the DNA that was purified - and let me remind you again - this purified DNA is the fraction of the genome that was in association with the protein of interest. Then you sequence it and you are able to identify which of the areas of the genome are the areas that were in association with your protein.
Just to summarise the differences between ChIP-chip and ChIP-sequencing, I took this table from a review by Peter Park, and I think I marked by stars the important points that I want to talk about. For instance, when I add this star the maximum resolution of an array is approximately 30 to 100 base pairs, and this is because you need the hybridisation to a specific probe. While for ChIP-sequencing you can get single-nucleotide resolutions, and people have begun using it also to identify single-nucleotide polymorphisms within ChIP-sequencing data. Next, another important point is the required amount of ChIP DNA where it's pretty high for the ChIP-chip, and for ChIP-sequencing we, and others, have managed to get it to low. This table is from 2009, and our field is moving so fast that a lot of advances have claimed that we can go to much lower than the numbers in this table. The dynamic range is also pretty limited in ChIP-chip, while it's not limited in ChIP-sequencing. You require amplification for ChIP-chip, but less amplification is required for ChIP-sequencing. Finally, multiplexing is possible only for ChIP-sequencing, and one of the studies that I will present later on shows the ability to use multiplexing in order to do high-throughput ChIP-sequencing.
I now want to present to you a small kind of quick look into the computational approach, to give you an idea of how it is done. As I mentioned earlier, after we derived the genomic areas, the DNA from the genomic areas that were in association with the protein of interest, we put them into a sequencing machine. This sequencing machine can generate millions of reads; I think nowadays you can get approximately 100 million reads per run in a very easy way, and you even have smaller machines that run overnight and can generate approximately 12 million. Then you take the reads and you need to figure out where do they go in the genome? There is a computational pipeline and this became more and more established since 2007 when it was first introduced, that align the reads and then you can either extend the reads because the assumption is that every read is derived from a nucleosome, and the piece of DNA wrapped around the nucleosome is usually approximately 150 base pairs.
So even though the reads that we get are only 35 to 40 base pairs, we can extend them after the alignment saying, 'Okay, we know that this read was derived from a fragment which is approximately 150 base pairs'. Then you can count how many fragments fall within each genomic position, and this allows to really get maps of the organisation of chromatin or the organisation of proteins that are in association with the DNA. Another way to look on the approaches for the analysis of ChIP-sequencing data, and, again, this was from a review from 2009 and there is a lot of advances nowadays, and I will revisit them later on, but I think this summarises really nice the approach that is going on. After you get - and I mark the stars where I go within this schematic. We start with a sequencing platform and then we also use a control sample, which we usually term as wholesale extract. This allows us to learn about the syndication procedure, the alignability of different genomic regions, whether they have repetitive areas or not? This, the wholesale extract, is what we do is we just take the cells that we use and extract the DNA out of them without pulling any proteins, so this is the background that we use to understand what is going on.
Then we do the alignment that I've described before, and the next step is what we call peak calling. There are numerous algorithms that came in the last few years that are able to detect genomic areas that have more enrichment for specific, for the signal above the background; some of them take into account the wholesale extract. The next step is now that you have the genomic regions that are enriched, is to visualise them. I will revisit later on two of the very nice tools for visualisation, IGV and the UCSC genome browser. Then you can do with this discovery that allows you to see if a specific genomic area is enriched for motifs that can tell you something about the binding of the protein. You can figure out relationship to the structure of the gene, and so on and so forth. This was my introduction and, again, if you have any questions please post them and I will revisit them by the end of the talk.
Now I want to talk about the description of genome-wide chromatin organisation, and a very nice study - or actually two studies, and I present to you only one by Ernst and Manolis, and the Bernstein Group as well - was used in order to identify the combination of histone modifications. But as this rogue data slide is pretty complex, I want to take you through a cartoon that puts the idea in a much easier way to understand. The basic idea is that when you look on the combination of histone modifications, and this could have been possible after a lot of the maps were generated. After people were able to take ChIP-chip and do ChIP-sequencing, they made a lot of maps for different histone modifications. What came out is that, so we got a lot of these maps and when you put them together you realise that the genome has discreet combinations of histone modifications. The study that I introduced before by Jason Ernst, has used a Hidden Markov model to break apart the genomic combinations of histone modifications.
What you can see here in a very nice way is what we call states, and, for instance, if this is a gene and TSS stands for Transcription Start Site, TTS stands for Transcription Termination Site. You can see that, for instance, in the transcription start site, which the area around it is what we call promoter when we talk about mammalian cells, you can see binding of RNA polymerase, but you can also see the variety of histone modifications that goes along with active promoter. For instance, as I mentioned earlier, H3K4me3, histone 3 on lysine 4 has three methyl moieties is usually marking the area of active genes. You can also see that it has other combinations of histone modifications, and when you look, for instance, within the gene you can see that at the beginning of the gene body there is H3K79 with 1, 2 or 3 methylation; and further into the gene you can see H3K36me3. What I want to tell you here is that this combination, when you take the different combinations of histone modifications this allows you to understand the way the genome is organised.
Another approach, and similar approach that was done in Drosophila used mapping of chromatin regulators, and in Drosophila they have identified five types of chromatin and they named them according to colours. Because, initially, chromatin comes from the Greek word 'chroma', and they also see that the organisation to the different chromatin colours goes along with the regulation of genes, and with the organisation of the chromatin within the nucleus. Another model that I want to present to you is the fine-tuning of genomic elements by histone modifications. What we have here is a dashboard of histone modifications, and this allows to somewhat - and, again, this is a model - fine-tuning of genomic elements. For instance, if we look on the left side we see that as promoters there could be either histone modifications that goes along with inactive, such as H3K9me3 or H3K27me3. There are combinations that goes along with poised, which are genes that are standing and waiting to be either activated or repressed; and having the dual combination of the active mark H3K4me3 and H3K27me3, as well as H2AZ. In the active area, we have a combination of acetylation on the histone tails, as well as H3K4me2, H3K4me3 and, again, H2AZ.
In a similar manner, we can see such modifications on enhancers or even within gene bodies. Finally, there are association of modification, of histone modifications such as the H3K9me3 and DNA methylation on the right side on the bottom, where you have stable repression usually goes along with this kind of modification. While a more transient repression is associated with H3K27me3 that is found or shown to be found, in some extent, to what we call polycomb bodies. This is, in general, the way that we envision the histone modifications to be organised and to allow the cells to precisely regulate their expression of genes. Another way to understand it is to get these high levels of fine-tuning, it seems that evolution has managed to devise a very sophisticated variability of what we call chromatin regulators. When I talk about chromatin regulators - and I will revisit this later - I talk about proteins and complexes that can modify or read, or write, or move the different component of chromatin. When I consider them, we see that it's very complex; there are about more than 100 isozymes, which are the same enzymes that can do the same catalytic reaction. Such as you have a variety of HDACs, which are histone deacetylases, you have a variety of HAT histone acetyltransferases and so on. Another important point, is that the chromatin regulators are tissue-specific and developmentally regulated. Finally, they have been shown to be implicated in cancer; so the cancer sometimes takes advantage of chromatin regulators in order to repress or activate genes that make it more, allow the kinds of cells to proliferate better.
Now, I want to present to you some genomic visualisation tools, and the way that we can use them in order to learn about the chromatin structure after we've managed to map it. The UCSC genome browser, one of the first ones to appear, has a combination of a lot of datasets and you can use it from downloading tables of genomic features, to find primers for PCR, and so on and so forth. I put on each of the slides that describe the tools, links to a much more detailed description of the tools and the way that you can use them. What I want to present to you here is the way that we can take ChIP-sequencing tracks and visualise them on the UCSC genome browser. One of the nice things here, is that what I did I took a map for H3K4me3 and you can see under it the variety of genes, you can add to it and remove according to your will there; boxes that you can check below in the page CpG islands, repeat elements, conservations and so on and so forth. For instance, when you get a dataset and you want to explore it, it's very good to use it in order to figure out which genomic regions are in association with the peaks or with the dips in your dataset.
Another very useful tool is the IGV which stands for Integrative Genomics Viewer. A very nice description that I took from Jim Robinson from the Broad Institute, is that this is a desktop publication for visualisation and interactive exploration of genomic databases. You can use it to visualise epigenomics, microarrays, RNA-seq, next-generation sequencing alignments and so on. One of the nice things about it is - and it's actually in the bottom of this list here - is that it provides a Google Maps-style interface, so you can scroll up and down in a very fast way, because it sits on your computer, it doesn't require going through the web. You can use it to visualise large datasets, and then it has a lot of diverse genomic data types that you can use it for.
Along with the fast-pace the field of genomics has been moving, there are many tools that were designed for the analysis and query of datasets. So, recently, there was introduced GenomeSpace, and the idea with GenomeSpace that it brings together a variety of computational tools that enable scientists, even with very limited programming skills, to easily combine the capabilities of different tools. People have generated very good tools to learn about the dataset, but it's sometimes very difficult to move from one tool to another. So this allows to have a common space that one can create and manipulate, or share the genomic data that you have. I think this is growing, but just to give you a feeling of what you have now it combines tools such as Cytoscape, Galaxy which is very good for analysis of ChIP-sequencing data, GenePattern, Genomica, IGV and the UCSC Genome Browser. A more detailed description of that is in the link below that has the entire slideshow from Michael Reich.
The last part of my talk, before I move forward to Miriam and then to your questions, is some major scientific discoveries that were stemming from the charting of in vivo chromatin maps. One of the very exciting discoveries or abilities that came together, is the way that we can now identify and characterise in much higher confidence, areas that we call enhancers. People realised that cells have enhancers many years ago, but it was very difficult to locate them because, as opposed to genes that are long and have very precise features, enhancers can be located from hundreds to even more kb upstream to the gene that they regulate. One of the very interesting points that is presented in these studies, is that when you consider, for instance, two cell types and you look on the gene expression, many of the genes that are expressed between two very different cell types are similar on the right, you see that approximately 8,500 are shared between the two cell types. But, on the left, what you see is that the enhancers predicted for these cell types are very distinct, and you can see that on the map on the right where the red marks, enhancers as mapped by H3K4me1 which is a histone modification that you usually associated with active enhancers.
So you can see that on the cells on the left and the HeLa cells on the left, a lot of the enhancers are unique to HeLa, while when you look on K562 it has also a very distinct set of enhancers. On the right side, it's more to show the complexity of the enhancers when you consider a variety of cell types, and you can also see that many of the enhancers have binding of transcription factors. Another point that I want to mention, is a study that Oren Ram and I presented about a year ago, and we tried to learn about chromatin regulators. As I mentioned before, there were many maps describing what is going on with the histone modification and the way they are organised, but little was known about the way chromatin regulators were organised. So what we did, we used an approach that allowed us to find antibodies that were good for ChIP-sequencing, and we screened many antibodies to get the one that works best for us. Then once we had the datasets we learned about the organisation of chromatin regulators, and we saw that chromatin regulators binding is combinatorial, is bifunctional, which means that we can see chromatin regulators that are activating or repressing, sitting together in the same area and that they are modular.
We have a very nice website for all the datasets that we have generated, and the next topic that I want to mention is that recently the ENCODE Consortium that I'm very proud to be a part of, came with a set of studies describing the exploration of genomic chromatin maps in human cells. I encourage you to visit the Nature ENCODE Explorer that allows you to choose a specific thread and see what studies were related to it in nature or in genome research, and this allows you to get a very deep understanding of the chromatin organisation. One of the points that I want to mention here, is together with DNA as one sensitivity it allows a very high resolution of characterisation of regulatory regions where you can identify the specific conservation and the specific motifs that are bound by transcription factors, and we see how they are related to the organisation of chromatin.
Another approach that I want to present here is a nice study that we presented recently, and worked on by Manuel, Nir, Ido and others, was the dynamic profiling of transcription factors. So we have developed a way to do ChIP-sequencing in the high-throughput manner, and what we use it for is to see the way the binding of transcription factors change after induction of, in a response of immune cells. Finally, I want to tell you about my studies and my future direction, so I'm interested in understanding the pre-implantation development and learning about the chromatin organisation and the patterns of chromatin in in vivo samples. We have been using methods that allow us to get ChIP-sequencing maps for a minimal number of cells, and then this allows us to figure out the developmental epigenomics and the way that chromatin is used to regulate the expression and pluripotency of cells during pre-implantation. Finally, I want to thank all the people that helped with the work, the ENCODE Consortium, EMBO, the Hood Foundation and the CEGs for their funding. I would like now to pass the talk to Miriam, and I encourage you to submit any questions you might have. Thank you for listening this far.
MF: Thank you very much Alon, that was a very interesting talk and I'm sure you will have plenty of questions from our listeners waiting for you. Hello, I would like to take this opportunity to tell you a bit more about some of the resources and products that Abcam has available for epigenetics research, with a special focus in ChIP. We have recently published the second edition of our general protocol and troubleshooting book, which contains general protocols on how to perform IHC, Flow Cytometry and, more interestingly for you, ChIP. You can find more specific clips and protocols in our epigenetics microsite at abcam.com/epigenetics, or check out our protocols page at abcam.com/protocols. You are probably aware that we have recently added two new applications, and our promise guarantee. RIP which stands for RNA Immunoprecipitation and CLIP, which stands for UV Cross-Linking and Immunoprecipitation. These are antibody-based techniques used to study RNA and protein interactions.
You can download the specific protocols for each of these applications at abcam.com/CLIPprotocol and abcam.com/RIPprotocol for CLIP and RIP, respectively. If you have any questions regarding the topics discussed today in the webinar or on any Abcam products, please feel free to contact our scientific support team who will be very happy to help you with any query you might have. For those of you who are located in the US, Canada or South America, please contact our US team. If you are in Hong Kong, China or Asia, please contact our Hong Kong team. If you are located in the UK or Europe, please contact our UK team. If you are in Japan, please contact our Japan team. I would like to highlight that we have multiple language support in German, French and Spanish, so please don't hesitate to contact us if you have any questions. As I've just mentioned, abcam.com/epigenetics is our epigenetics microsite, a one-stop shop for all topics relating to epigenetics. In this microsite you can find the latest information on products, protocols and Abcam upcoming meetings.
If you use large quantities of one product, you might one to look into buying in bulk. Bulk-buying will help you to save money and minimise the viability in your experiment. For more information about our bulk options, please contact our sales team at Abcam at email@example.com. I would like now to focus on Abcam's range of epigenetic kits, the EpiSeeker. This kit has been designed with researches in mind so that you can spend less time doing the experiment, and more time thinking about the design and the outcome you want to achieve. Although the EpiSeeker range includes a lot of these end products such as histone modifications or methylation quantification kits, I would like to focus today on our range of ChIP kits. All EpiSeeker ChIP kits are for cross-link ChIP, and therefore cannot be used for a native ChIP. Our one-step and plant ChIP kits have been optimised for mammalian and plant DNA, respectively. These kits don't contain a preselected antibody, and therefore can be adaptable to any target of your choice. The EpiSeeker range also includes kits optimised for methylated and acetylated histone modifications, like the ones that Alon has mentioned in the webinar, and can be used either for cell or tissue starting material.
We also have several kits for immunoprecipitation of methylated DNA on which I will elaborate a bit further later on. More details on these kits and other EpiSeeker products can be found at abcam.com/EpiSeeker. You might probably ask what are the advantages of using EpiSeeker ChIP kits instead of following the conventional ChIP method? Well, the reaction takes place on a 96 well plate, so they are easier to standardise. It only takes five hours, as opposed to the normal method which generally takes about two days. The kits contain all the main reagents, except the formaldehyde for the cross-linking step, and except for the general kits which I've mentioned in one-step and the plant ChIP. The kits contain a preselected ChIP grade antibody which has been optimised for the assay. Moreover, the precipitated DNA can be used straightaway for downstream processes, such as the ChIP-chip or the ChIP-seq, which have been discussed by Alon in this webinar.
As I previously mentioned, we do offer as well methylated DNA immunoprecipitation kits. These kits contain specific antibodies to a specifically enriched methylated and hydroxymethylated DNA, respectively. These modifications seem to play an important role on differential gene expression, and therefore it is important to have proper tools to investigate their function. Our immunoprecipitation kits are also useful to complement other DNA methylation experiments, such as cytosine modification where it's not possible to differentiate between methylated and hydroxymethylated cytosines. As a thank you for attending the webinar we are offering you a 35 per cent discount on any ChIP kit, and this offer is valid for all EpiSeeker ChIP kits as well as for our standard ChIP kit Ab500. After this meeting you will be directed to a website where you can find more information about this special offer, plus the downloadable copy of this webinar.
I would just like to highlight Abcam's Chromatin, Replication and Chromosomal Stability meeting that is taking place in Copenhagen next June, as this might be of interest to some of you. If you would like more information about this meeting, please visit the meeting website at abcam.com/Copenhagen2013. Without further delay, I'll pass you over to Alon who is ready to answer the questions that we've been receiving during the webinar. Thank you very much for your attention.
AG: Thank you, Miriam. I will try to go over the questions and answer them to the details. Monica asked: What is the main difference between nucleosomes and histones? As you can see in the beginning of the talk, the nucleosomes are composed of histones, so the basic idea that each nucleosome which is this barrel-shaped complex is composed of eight histones. The histones themselves are - they also vary within the nucleosomes, and they have different modifications, so this allows to have more complexity in terms of gene regulation. Monica also asked if there is a situation where ChIP-chip is a better technique than ChIP-sequencing? To some extent, for instance, in creatures, in organisms with not so complex genomes such as Drosophila in yeast, sometimes it's easier to do ChIP-chip because you can tell the entire genome. Some of the ways that people are using arrays, which this is the ChIP, the second ChIP in the ChIP-chip, the arrays that have long probes on them, some of the methods that people are using today is to fish or to generate sequences of long oligos in a high-throughput manner, and then to either probe genomic regions for sequencing or to deplete genomic regions for sequencing. This way, when you go through the sequencing procedure, the parts of the genome that you sequence are more specific according to what you want to focus on. For instance, a lot of the exsome mapping are using such an approach.
Susan asks: How do you choose endogenous control for real-time PCR or ChIP array, given that we've already selected only for proteins bound to the target? In general, and I think this is one of the terrific advantages of the area that we're in now, is that there so many ChIP-sequencing maps available that if you want to focus on some kind of histone modifications, it's probably feasible that you will find a cell-type similar to what you're looking for, and find the maps for it and then you can look on the genomic organisation and design primers for the area of interest. In terms of the background, what we usually do is we choose more than one because you want not to rely only on one genomic area, and not on one set of primers. Then once we use this background it is usually from genomic regions that we expect them to be devoid of the histone modification. Another approach is also to take areas that have high redundancy. For instance, you can take some repetitive area of the genome and expect it to be enriched to similar levels within the sample and the ChIP itself, and this way you can normalise to this background.
When we did the chromatin regulator study one of our main problems was that we didn't know where to expect chromatin regulators to bind. In order to fish them, we used an approach that was built upon the chromatin states and we made a variety of probes that recognises each chromatin state, and we assumed that chromatin regulars in general will be binding to chromatin states. Doris asked: What is the way to check the quality of the DNA after we do ChIP, and before we do the library preparation? If you start with a high number of cells and you can measure the amount of DNA, and you can run a PCR on some genomic regions that you expect will have enrichment, and you expect will not have enrichment and this way see if you get enrichment that you expect. But if you are very limited with the amounts that you have, it's sometimes better to do these quality checks after you've made the library, but before you sequence it. So you can do the quality check on the library which, to some extent, amplifies your sample in a very good way that has less bias than, for instance, whole genome amplification.
Another question is what marks do we look for in enhanced regions? In general, people have seen that H3K4me1 and H3K27 acetyl are associated with active enhancers. Another point to add to that, is that you would expect that enhancers will not be located within promoter regions. So if we want to find enhancers, we would usually exclude peaks of these marks within promoter regions. Another question is about looking for a software for preliminary analysis of ChIP-sequencing data. I would recommend to look into Galaxy and, again, I think that a lot of the ease of use that comes from GenomeSpace can be very useful to many people. One of the great ideas in GenomeSpace is that you upload your data once into the crowd, and then you can just move it between the different tools and visualise gene expression, and visualise analysis of your ChIP-sequencing data using Galaxy, and visualise how it looks on the genome viewer and so on. So I think on this and the GenomeSpace is a very good way to start.
A question is how do we carry amplification after a ChIP, and which primer we're using? This depends on the method used for exploring your DNA later on. We routinely use the Illumina sequencer and we use their library preparation, but I know that they have a variety of methods and they depend on if you go for sequencing, or if you go for ChIP-chip and so on. Tanya asks: What is my opinion on native ChIP-seq versus cross-linked ChIP-sequencing? The good part of native ChIP-seq is that it allows you to look at genomic areas that are in association with the protein of interest, or with nucleosomes that have a histone modified at the tail with a point of interest. In a way, that is very, very rigorous, which means that you keep the nucleosomes on place only by maintaining the chromatin in proper buffer and proper temperature. The issue that can come with that is that it's more sensitive, and if, for instance, the buffer or the conditions make the nucleosomes move, it can give you a different picture than what happened when you harvested the cells. What I like about the cross-linking is that it allows you to get a snapshot of what was going on when we added the cross-linker. But there are a variety of studies trying to look into the differences between them, and I think for some approaches native ChIP-seq is better than cross-linked than the other way around.
Again, I want to thank you all for listening and if you have any other questions that I didn't answer due to the time limits, we will contact you within the 48 hours with an answer. Thank you very much and hope you have a great day.