Contact Us

Costa Rica concentrates 5% of the planet’s biodiversity. However, there is little information at present based on this reality; and this is just one example of the situation in Central America and the Caribbean. Here is where the need to have the Central American Bioinformatics Network becomes palpable. We talked about this with its leader, the expert in bioinformatics and molecular nanotechnology and currently general manager of Spain’s National Institute of Bioinformatics (Madrid), Allan Orozco.

When was the Central American Bioinformatics Network created and what were the objectives?

In June 2011 with the creation and endorsement of a document (MoU) signed by scientific representatives from several countries like Guatemala, Costa Rica and Panama. The Central American isthmus is an area with limited economic resources and, therefore, it is convenient and advisable to start with an “in silico”, an activity that’s cheaper than directly doing experimental genomics; the scalability of the process is simply a matter of time. On the other hand, the key objectives of this specialised network are: To develop and implement research in molecular medicine; to analyse and process biological information in HPC (high performance computing) in a cooperative exchange, storage and processing network; to develop converging and peripheral areas such as computer nanogenomics and space control in parallel dynamic models by means of comparative genomics; to encourage the creation of bioinformatics cloud biocomputing on NSG and virtual machines, etc.

What was the initial motivation?

At present, we are proposing an organisation model based on seven interconnected nodes and a central node in Costa Rica. As regards motivation, I think it is important to create a sustainable enough base in biocomputing for the benefit of Central America and nearby zones like the Caribbean.

This initial motivation stems from the general interest in improving genomics analyses by means of biocomputing in the Central American isthmus. In Costa Rica we have the Nelly cluster, with 60 servers and a capacity of up to 100 TB of distributed memory, the largest in Central America and the Caribbean exclusively dedicated to processing biological data in Bioinformatics. The processing of larger data is backed by the Calendula cluster at the Supercomputing Centre in León, Spain, which is connected to RedIRIS, which is in turn interconnected to RedCLARA.

Can you explain a bit further the role of RedCLARA within the Central American Bioinformatics Network?

The Nelly cluster is one of the engines of the Central American Bioinformatics Network and is connected to RedCLARA from the Informatics Centre at the University of Costa Rica; this has been a key factor to support all the applied informatics activities in Costa Rica. It is important to mention that the cluster installation has become a good example of triangular cooperation between a private enterprise, a university and the Government. It is vital to consider technological transfer towards the entrepreneurial sector from the academia is essential in times of crisis, cooperation being an essentially valid effort, between operational frameworks and intelligent dynamics agreed on the basis of collaborative strategic actions. We cannot continue to do academic research with little social applicability and a low-impact profile for the different productive cells in the country. This is something we must change in terms of attitude, frame of mind and will.

On the other hand, RedIRIS is a high capacity grid network in Spain, with nearly 18 nodes distributed across its territory; RedCLARA interconnects America with Europe through a link in RedIRIS which connects it directly to GÈANT (Pan European research and education network with 12.000 kilometres of optical fibre and photonic service and IP, LightPaths, etc). Both networks currently have a second version in an implementation stage. RedCLARA is simply the interconnection media for universities and research centres at a continental level between Europe and America.

Are the data which arrive in Calendula sent from Central America through the RedCLARA connection to Europe or is this not working like that yet?

The operability and configuration can be relative and adjustable to each case. The normal protocol is RedCLARA – RedIRIS –GÉANT and vice versa (depending on which critical route is in operational transit and PoP nodes in station band). Transmission times are used by the interconnectivity of the distributed point networks in Spain and by both continental grids. The entry point is in TELVENT (housing mode) in Madrid and, in fact, both the GÉANT and the RedIRIS nodes are in the same location. The Calendula cluster is connected to RedIRIS and opens the connection doors towards America. With the extensions of RedIRIS-NOVA (created in 2011) in dark fibre, it is possible to reach a speed of up to 10 Gbps per hour, which can be expanded up to 100 Gbps or more.

In Spain, Calendula expands Nelly’s operational capacities in Central America. RedIRIS is an academic network; whereas ours is oriented to the provision of web services to carry out specific tasks in bioinformatics projects. This will favour, for example, cloud web services to integrate several functional frameworks and reusable genetic records applied in cases like Central America’s biodiversity, one of the most important in the world. Only in Costa Rica we have 5% of the world’s biodiversity. Bioinformatics exploration in biodiversity has been very poor in Central America and the rest of America; this will have to change with the actions and the work of our Central American Bioinformatics Network.

Is the network already structured and are the nodes interconnected? Or which of them are currently interconnected and when will the others be interconnected?

The objectives of the technical plan are being met and therefore I think that by the end of 2012 we will have an interconnected grid between Central American countries which is stable enough to migrate and expand interconnections towards additional routes in North and South America (in the future I think RedCLARA should check the conglomerate of islands in the Caribbean). Costa Rica does not have a PoP inside RedCLARA, but we must take advantage of the fact that there is a point in one Central American country: Panama.

Only four countries in the Central American Bioinformatics Network are connected to RedCLARA at present; do they use the interconnection between them through RedCLARA to operate or are they not operating yet?

Actually, at present only Guatemala, Panama, El Salvador and Costa Rica are connected to RedCLARA, which is a valid transit path. We must make an effort so that Honduras, Belize and Nicaragua join the same initiative in order to complete the zone in its entirety in terms of nations.

In relation to these three countries, can a network with a high demand of data processing, like the Central American Bioinformatics one, operate partially or totally over commercial internet? Have you considered the possibility of establishing dedicated connection points in order to incorporate those three countries? Or will you operate in a way which combines the capacities of RedCLARA with those of commercial internet?

We have been talking to representatives from the AMI Network in order to compensate this situation, maybe under the concept of hired transmission capacity. Apart from its electrical infrastructure it has a communications thread which goes from Mexico to Panama of which we can take advantage to solve the problems identified. We have to solve the differences by means of an agreement and negotiable stages, and we do not rule out the possibility of operating under a complementary commercial traffic even though we are aware of its constraints in terms of capacity and the protection of genomics data and all the related security. This issue will be highly important in the future, when the costs of genomics sequencing become significantly lower.

For example we must remember that RedIRIS-NOVA (a project of more than 90 million Euros) has a maximum speed of 100 Gbps, around 10.000 greater than domestic internet connections (10 Gbps). Therefore, there is an evident competitive advantage in using these networks in research activities as opposed to classic networks, especially in terms of the distribution and transmission of large volumes of biological information.

I understand Guatemala’s Bioinformatics Network is now connected to RedCLARA, within the context of the Central American Bioinformatics Network. Which activities are being developed today thanks to that connection from Guatemala?

This is fascinating, because based on the central node we have carried out support initiatives on the constitution of local networks in each country, as in the concrete case of Guatemala and Costa Rica. Maybe it is not the ideal way of starting, but it is a functional reengineering which has been very well received in the region. In Guatemala we recently introduced within the network, to SENACYT, a project called Strengthening of the use of applied bioinformatics in genetics which contributes to increasing Guatemala’s agricultural competitiveness. This will make it possible to build a specialised bioinformatics network with switched nodes with application in the agricultural field in Guatemala. We have to wait for the resolution, but we are very optimistic about this; it will allow us to expand and develop related tasks in complementation with the Bioinformatics Network.

How does RedCLARA represent a contribution to the Central American Bioinformatics Network? What are you currently doing which you could not do if the connection to RedCLARA didn’t exist and what are you planning to do in the future thanks to this connection?

The contribution of infrastructure is the first, and the second is the complexity of web systems (e.g. the case of GISELA –virtual grid communities as part of FP7- a European project associated to RedCLARA) which we could install through the virtualisation of computers in a cloud and bioinformatics systems like the GALAXY project. The security and integrity of genetic data is a topic that we have to carefully analyse and we must have the support of biosecurity.

One of the ways in which it is useful because of its nature is the development of a virtual continental platform for bioinformatics (building on the Central American bioinformatics network’s base). This will make it possible to develop educational transfer in the field of bioinformatics and computer biology, for instance in topics with greater demand, such as the analysis of genomic data in the New Generation of Sequencing (NGS).

What are the tasks for the present and for the next three years for the Central American Bioinformatics Network?

Fundamentally, we plan to consolidate its integration and functioning in the Central America and the Caribbean region. Subsequently, we plan to open derivations without congestion in North and South America, in order to turn it into the next American Bioinformatics Infrastructure.

What is the relevance that you attribute to RedCLARA for the region?

It is important to start incorporating valuable projects with divided themes. The traceability of information processing under cooperative schemes of mutual collaboration can provide a lot of profits in joint work.