international initiatives and I will in my talk explain what happens at a more local level, at a regional level. So I will talk at first about our network

of core facilities, about what's happening in biology and then after that I will talk about our project called e-bio-genoist whose aim is to bridge data, metadata and computation. And let's go. So bio-bio-genoist, as we pronounce it,

is a network of core facilities and it's in two regions, Bretony and Peil-Elwar. And this network was created in quite many years now and it coordinates 31

technological core facilities in the two regions, Bretony and Peil-Elwar. And it's funny to see that Peil-Elwar cannot be translated in English but Bretony can be. And the aim of this network is to federate and organize the actions of the

different institutes because it will also federate more than six 70 research units and that for many topics, many scientific topics, mostly marine science, agri-food, health and bioinformatics. So Genoist, you will

hear many waste things, the far-waste syndrome I guess. Genoist is a bioinformatics core facility. We're like a bunch of Boy Scouts. We like to collect badges and

labels. So we're obviously a member of the Bio-Genoist network. We're also a member of the EFB, the French Bioinformatics Institute. And we're also recognized at the national level by IBISAR. We're also recognized by our colleagues from INRA as a regional strategic facility and we're also ISO

certified. As a platform, as a core facility, we provide typical things that any core facility should provide such as computing infrastructure,

storage, software development, expertise, and we are carrying out also many different projects around the different challenges that we are facing right now. Among our projects, we have many different focuses. Among them,

we will find computation, workflows, portals, collaboration, data, metadata, and ontologies. For example, in computation, we are trying to find the

best way to bring computational power to solve the problems that biology is encountering. So we worked on grid computing. We are now working

also on cloud computing. We are operating one of the two academic clouds dedicated to bioinformatics. We are also heavily involved in the usage of utilization of portals such as mobile or galaxy, mentioned earlier. And

with all these projects, we can, we were able to launch what was called e-biogenoist. It's an e-science initiative for the life science in the

West of France. So the context is the following. If you see the cost of the e-biogenoist since 2008. And this is a great thing for biology because we

scientists were able to address problems that could not be addressed formally. But it has a very strong implication because the decrease of the cost allows this community to launch more and more sequencing programs. And

it's very common to see, to meet a biologist who come with his hard drive and say, okay, I had some remaining credits and I was able to sequence my

organism or many of my organisms. What can I do now? The problem is here. They are able to generate massive amounts of data, but very often the infrastructure is not here. The IT infrastructure is not here. They have

not enough people to analyze the data. And typically this is a real bottleneck. And the situation can be worsened by the fact that the speed at which the data are generated is now outpacing the traditional Moore law that

describes the progress in the computational power. So in this case you can see the two curves describing the speed of the microprocessor and the storage capacity. And you can see clearly that the curve of the

sequencing capabilities is completely crushing everything. And in this case, we are just focusing on the genomic field, but there will be another wave

of data such for the proteomics, for the bioimaging, that will also cause very dramatic changes in the way biologists are doing their research. Typically,

we can consider that biology has now turned as a digital science and generating huge amounts of data, heterogeneous amounts of data. And this situation can be critical for some laboratories because this evolution was not clearly anticipated. It appeared very recently, the first next-generation sequencing

machines appeared in 2005. So in less than 10 years, discipline has to restructure his whole organization. Focus mostly on the generation of data. Now they must reorganize themselves to be able to analyze the huge amounts of

data they are generating. And that's where we come. Our project was started in May 2012, and it's funded for three years. It's quite a modest

funding because we have one person dedicated for this project, funded by the two regions. And what we're trying to do is to do some community buildings around the challenges that biology will have to face. So we're

providing many training workshops about this issue. And we're also preparing a roadmap to help the whole network to build coherent infrastructure to deal

with this amount of data. And in this project, we're also building a pilot project that can become our prototype prototypal virtual research environment. How have we built this system? Just by building a

system of systems. Because the resources were quite limited, so we chose to aggregate different tools to be able to cope with the data integration and analysis problem. So we combine different tools such as

Galaxy that was mentioned earlier. We also used ISA tools, the ISA tools suite. And also a collaborative portal such as Hub Zero. Of course, we had to use additional software to make the whole thing work using some APIs,

doing some developments, or using some dedicated tools. The Galaxy is a web portal for biomedical data analysis, and it's widely used now. There's a

tremendous success of this tool because mostly of this intuitive interface, and also because it allows the user to build workflows, and these workflows can be exchanged between users. So there's a very strong

momentum around this tool. And we also have obviously an instance of Galaxy at Gen West. And on this, we have inserted 800 tools covering different fields of different topics of biological data analysis to help

our users. We also use ISA tools suite. So it's a tool dedicated for the management of experimental metadata. It was created by people at

Oxford, and it allows people to describe from a technical point of view their experiments. It can be considered as an electronic lab book,

for example. When the description of the experiment is done, there's a strong encouragement through the interface to use ontologies to have a better vocabulary. And it will also help people to create local

repositories where they can centralize all their data and they can share their experiments. And of course, it allows also publication to public

repositories. It's one of the main advantages. There's some cost of describing, of taking time to describe the experiment, and the reward is the fact that there will be some easier publication steps after that. And we

also at Gen West have used the ISA tool suite, and we have also have adapted this suite to our needs. And we also made some additional developments for a better adaptation. So what's happening exactly? So the wet lab experiment generates data and the metagenics data

and can be described with the metadata. And with ISA tools, you will have the ISA tab files that can link to raw data. And with the ISA tools,

you can also be able to create an archive, the ISA archive. And this archive, we developed different tools to import the ISA archive in the Galaxy environment. In this case, the data are imported,

and the Galaxy environment is now aware of the location of the raw data. And in this case, people are able, in a very simple way, to do their analysis quite

immediately. So once they describe the experiment, they explain where the raw data are located, and then the next step of the Galaxy import allows them to

launch their own analysis. The next tool is the HUB0. It's a scientific web portal that is widely used in the nanoscience community, for example. It gives many

functionalities for the collaboration, such as Wiki, blogs, and everything is built around resources, such as results, articles, presentations. And there are also some modules to do some lightweight project management. And combining these three tools,

we are able to provide a single integrated environment for the management and the analysis of the biological data, and with a collaborative resource. That creates something that could be

called a virtual environment, virtual research environment, with the web portal. Everything is built upon a computing infrastructure, hosting, of course, a data infrastructure. There are the functionalities provided by HUB0 that help the scientists to

deal with the different steps of the life science project of the project life cycle, project management, collaboration, dissemination. And so with the data,

the scientist is able to launch very easily its workflows to analyze the data. So what we're trying to do here, and it's exactly what's happening everywhere, I guess,

is we're changing the paradigm, because formerly biologists came with their data, and we had to say to them, okay, put there on the Linux box, and you will find some software and try to parallelize your computation, learn to use a grid manager and everything.

So now, since the data are becoming more and more huge, and the size of the virtual machine,

as said earlier, is becoming, not becoming smaller, but it's rather small compared to the size of the data, we're now plugging the IT tools, the IT environment, onto the data. And it's a very important step in my mind, because when the first

side of the older way of working, we were de-processing the biologists from its own data. When someone has to find the good software in a huge list, when someone has to deal with

file formats, when someone has to deal with a new environment, when someone has to deal with the parallelization of some tools, he can easily lose the sight, lose the right purpose of his work. The biologist is here to answer

biological questions, not to lose his time doing some fancy computer science stuff. And in this case, I guess we could have a solution, a viable solution, to help biologists

to regain their data. So the next step, what we learned, so when you propose a new tool, some new methodologies, there's always the acceptance and the adoption issues.

It's incredible. Sometimes people say, okay, that's fine, I understand the benefits, but it will interfere too much with my way of working. So it's a rejection.

It can happen very often. It's better to deal with people who start new projects with no established rules. And in this case, they will adopt more easily the tools. For the moment, on the hub, we have more than 150 people trying, and we're hosting 50 projects.

The benefits are more obvious when the tools are used by different communities,

for example, biologists and computer scientists. And on this hub, they find the perfect tool to exchange very easily. So what we will do next, we obviously have to switch to a production

environment, production environment, because it's only a prototype for the moment. We will also investigate identity federation to cope with the multi-organism apartments of the people using our tools. And we're also working on the metadata for

bioinformatics workflows. It can be itself a protocol, and that we would like to grasp and to collect information about this type of experiment in silico. So what we need to do,

we have to connect to other initiatives, because we started at a very local level, and when I see and when I heard all the very nice projects during this conference, I know that

I have to reach other people. And what we need to do also is to define the perimeter. For bioinformatics facility, it's not an easy task to do. We are focusing mostly on computational and high-performance computation, and we must add some extra functionalities

as data management. And that's not what we can do at the best, obviously. So as a conclusion, biology has become a digital science. It's a new event. Very rapidly,

and the changes are very profound, and the usage must evolve also very rapidly. But for the moment, it could be a dangerous situation. If there were so many

bioinformaticians in France, it could be not so dangerous. But for the moment, maybe there are not enough people. So we tried to build a system of systems using recognized tools to build a better tool, and we tried to create a continuum

of data to bring back biology to the biologists.