The non-coding Arabidopsis genome

点击翻译按钮获取中文摘要

What we used to think goes something like this: Our chromosomes are made of DNA and embedded within them are our genes. When our genes are switched on, they are converted into a related molecule called RNA, which eventually converts the code of our genes into proteins that carry out all the tasks in our cells. Much of the DNA in our chromosomes doesn't encode such genes and is therefore junk. But we think of things a little differently now: Although the human genome was sequenced 10 years ago we are still working out what it codes for. To our surprise much more of it is copied into RNA than that which codes for proteins. We call these non-coding RNAs and it turns out that many play critical roles in controlling how protein coding genes are switched on, off, or modified. We now realize that to understand how genes and genomes work, we need to discover non-coding RNAs and work out what they do. My lab is interested in how plants control flower development. We have discovered that non-coding RNAs control when flowers are made and this has got us interested in non-coding RNAs. DNA sequencing has been revolutionized recently by so-called next generation sequence technology. This approach can be used to sequence copies of RNA too, allowing us to identify all the RNAs made in a cell. Copies of RNA are made by something called RT, so we almost never look at RNA directly. Unfortunately RT can make mistakes so our interpretations about RNA can be wrong. Last year a company called Helicos Biosciences developed a 3rd Generation sequencing technology that can sequence RNA directly. We have collaborated with this company to sequence the ends of RNA molecules in Arabidopsis (the first plant to have its entire genome sequenced). While doing this we also found lots of non-coding RNAs and lots of problems with previous experiments that used RT to look at RNA. In this proposal we want to build on our breakthroughs with this technology to discover and annotate the non-coding RNAs of Arabidopsis. As the annotation of all other plant genomes largely derives from Arabidopsis, this will be invaluable in understanding other plant genomes, including the crops that we depend on for food and fuel. We will look for hidden RNAs and long non-coding RNAs. Some non-coding RNAs are destroyed almost as quickly as they are made, so they are normally effectively hidden. But they can be seen in plants that lack the exosome - a group of proteins that destroy RNAs. We have already shown there are lots of mistakes in a previous attempt to look at these RNAs, so we know that our approach is a great way to show these RNAs in accurate detail. The RNA sequencing we have already carried out identified lots of RNAs not previously found before. The trouble is our sequencing only identified the ends of these RNAs and we don't know the nature of the whole RNA molecule. A recent study of human non-coding RNAs found it helpful to fragment RNA, sequence, and compare special marks on chromosomes found where genes start and where the 'body' of a gene is. We will take the same approach here with Arabidopsis, but sequence RNA directly rather than RT made copies. In this proposal, we will also try to identify a function for a large number of the RNAs we find. In humans, long non-coding RNAs act as a scaffold for special proteins that switch genes off. We will find RNAs that bind to these proteins in Arabidopsis. My lab has special expertise here, as we were among the first to show how to do this in Arabidopsis. Finally, our recent breakthroughs identified a set of non-coding RNAs that mostly do a quite well worked out job: snoRNAs help modify other RNAs. Unfortunately, we found lots of problems in the way people have looked at Arabidopsis non-coding RNAs that can be explained by snoRNAs and RT mistakes. To prevent this happening in the future, we will sort out the annotation of these RNAs.