SB analyzed the data. GP, VG, and SB wrote the paper.
Sebastian Bassi is with the Universidad Nacional de Quilmes, Buenos Aires, Argentina. Virginia González is part of the Structural Bioinformatics Group at the Universidad Nacional de Quilmes, Buenos Aires, Argentina. Gustavo Parisi is the Structural Bioinformatics Group leader and professor at the Universidad Nacional de Quilmes, Buenos Aires, Argentina. He is also a researcher at the Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET).
The authors have declared that no competing interests exist.
Computational biology is an interdisciplinary science bred from fields as disparate as mathematics, chemistry, statistics, physics, biology, and computer science. Although the exact definition of computational biology is far from being precise and unambiguous, it is a fact that hundreds of scientists around the world have been increasingly using skills from the above-mentioned fields to approach different biological questions. It is not our aim to elucidate here the definition of computational biology and its differences from related fields such as bioinformatics. However, to review the state of this discipline in Argentina, we need at least a working definition. In this sense, any interdisciplinary research in which the main interest is in studying biological problems and where the working hypothesis can be tested by means of simulation and computational modeling will be considered as belonging to the computational biology field or at least as employing a computational biology approach.
It is interesting to note that computational biology research can be implemented in two different ways depending on how the “interdisciplinary” nature of this field is assembled. On one hand, a scientist with formal training in a given field (for example, a molecular biologist) could acquire other skills (such as programming or mathematical training) in his attempt to answer a given biological question. Alternatively, working teams made up of members specializing in different fields may work together to reach a scientific explanation of a problem. We think that the first approach is more common among the scientific community because it depends entirely on the scientist's desire to discover an explanation for their problems and on their capacity to explore different areas of science. The second approach is generally dependent on the availability of research and development programs, funded by public or private resources, that favor collaborations between different research institutions to form multidisciplinary teams. We will see that the first approach is more common in Argentina.
This paper summarizes the state of the art of computational biology in Argentina. Our aim is to offer as broad a view as possible of the different groups of scientists and their main research interests. Also, we present a brief review of educational, research, and development policies related to the field. We hope that this review may encourage overseas researchers to contact and collaborate with Argentinean teams, as well as organizing and facilitating the exchange of information between researchers in our country. However, this review will probably be far from complete, due to the lack of centralized information on computational biology, and for this reason we apologize for any potential omissions. (For author information, see
Based on the academic background of principal team members, we can differentiate two main types of computational biology groups in Argentina. On one hand, there are research groups coming from areas such as chemistry and physics. In general, these groups have a solid background in mathematics and statistics and they can easily adapt or develop computational techniques. Several of these groups have directed their interest toward biological problems in the last few years. Such groups generally offer atomic or molecular explanations using molecular modeling techniques based in quantum, classical, or hybrid quantum/classical mechanics. Also, and expanding beyond the molecular level, we found groups engaged in modeling metabolisms, signaling pathways, and simulating learning processes or auditory physiology. Some teams work in close collaboration with experimental groups to test their predictions.
Among these types of groups, we find the group headed by Adriana Pierini at the Departamento de Química Organica, Universidad Nacional de Córdoba. This group has expertise in the application of molecular modeling techniques, based on electron-structure methods, to the study of different mechanisms of relevance in physical–organic chemistry. These studies have been mainly focused on modeling reactions that occur through single electron transfer and are currently used to model oxidation reactions in peroxidases [
Dario Estrin's group focuses on the study of different processes in hemoproteins using computational classical molecular dynamics and hybrid quantum/classical techniques [
There are several groups in the Departamento de Fisica, Facultad de Ciencias Exactas, Universidad Nacional de Buenos Aires who are using computational tools to understand biological problems. More specifically, one of these groups, headed by Silvina Ponce Dawson, works on the mathematical modeling of intracellular calcium signals and on the development of novel numerical techniques to infer quantitative information from microscopy images [
The other type of computational biology research groups in Argentina consists of teams derived from biological disciplines such as biology, biochemistry, and agriculture. Many of these groups are involved in diverse genome sequencing projects and in the corresponding databases' curation. Most of the organisms involved in these studies are relevant to local health or agricultural issues. Other groups are involved in the development of bioinformatics tools for the characterization of sequence and protein structure or for the analysis of microarrays, while others are working in evolutionary studies and the development of molecular models of evolution.
At the Instituto de Investigaciones Biotecnológicas at the Universidad de San Martín, one group has been involved in the sequencing of the
In the Instituto Nacional de Tecnología Agropecuaria (INTA), there are several groups working in genomics and microarray data analysis [
Another team involved in genomic analysis is the recently established Centro Regional de Estudios Genómicos (CREG), supported by the Universidad Nacional de La Plata. They have a bioinformatics group, but are focused mostly on information theory and postgenomic analysis. The CREG was part of the sequencing effort of the
Information about Groups Working on Computational Biology and Bioinformatics in Argentina
A team devoted to microarray data analysis and genome analysis [
There are at least four groups working on structural bioinformatics of proteins. Rosana Chehín at the Instituto Superior de Investigaciones Biológicas of the University of Tucumán and CONICET has recently developed a method for clustering large protein families and subfamilies based in a linear array of conserved motifs and domains [
Adrian Turjanski, at the Departamento de Química Inorgánica, Analítica y Química Física, Facultad de Ciencias Exactas y Naturales, Universidad Nacional de Buenos Aires, focused his research on developing tools using molecular dynamics, quantum mechanics, and molecular mechanics and docking. He has recently been studying the underlying mechanism of phosphorylation and its regulatory role in mitogen-activated protein kinases [
The group headed by Elmer Fernández, at the Universidad Católica de Córdoba, mainly focused their research in proteomic and genomic data mining. Also, they perform studies in microarray data analysis, 2-D gel analysis, and statistical biological modeling [
As we mentioned above, scientists who want to work in computational biology should complement their basic degree training taking undergraduate or postgraduate courses. The path to do this is very far from being linear, and the lack of organized information about the minimum necessary knowledge to work in computational biology is very discouraging to new graduates from different areas. Moreover, the lack of information about computational approaches to studying biological systems and also about scientists performing this class of research in Argentina reduces the academic opportunity for new graduates to work in this field. However, there are a few exceptions, such as the degree in bioinformatics launched in 2006 at Universidad Nacional de Entre Ríos (
Although several universities offer biotechnology or biology degrees, very few include in their curricula any computational biology subjects. Universidad Nacional de Quilmes has offered Bioinformatics as an optional course for more than six years, and very recently opened a short specialization on programming. Universidad Nacional de Tucumán offers an Informatics course and others, and Universidad de Buenos Aires offers Introduction to Computational Biology (
The main problem is that most universities are organized into old French-style faculties—i.e., they offer a fixed or somehow limited choice of subjects for each degree. This scheme is used by the most important universities of Argentina. Universities structured into a departmental system integrate their subjects among the different degrees, as is done by Universidad Nacional del Sur and Universidad Nacional de Quilmes.
Most of the aforementioned research teams have developed their own software or adapted source codes from other authors, but in most cases for internal use only. Very few authors have released those programs for public distribution. We think that the main reason for the limited release of software in Argentina is probably related to the deficient computing and programming training of the average Argentinean scientist in the field, as well as to the lack of truly multidisciplinary teams. Also, the universities are, in the best cases, only just starting to discuss licensing policies to release software.
Among the very few examples of public software, we can mention the Linux distribution for bioinformatics called DNALinux [
Another system that deserves to be highlighted is wEMBOSS, a Web interface and data manager for the popular EMBOSS software package for biological sequence analysis [
Another scientific software program that has been recently released is a program that simulates sequence divergence under structure conservation of protein during evolution (SCPE from structurally constrained protein evolutionary model), developed by Gustavo Parisi and Julián Echave [
Scientific research is mostly restricted to governmental institutions [
Agriculture is one of the most prominent industries in Argentina. For this reason, most international biotechnological companies with agricultural interests have one of their branches located in our country. Several of these companies have their own bioinformatics groups, but they are small and their results are kept for internal use.
The emerging picture of the state of computational biology research in Argentina is characterized by the occurrence of fairly new groups, with limited or no interaction between them. Most of these groups were probably built around the insight of a scientist who had incorporated new disciplines and skills to open new lines of research. Finally, very few of the described groups are truly functioning as multidisciplinary teams. Fortunately, and considering that the critical mass of researchers in the field in Argentina has not yet been reached, Argentinean research projects have explored living systems at multiple organizational levels. We described groups focused on atomic or molecular analysis, sequence, evolution, structure, and dynamics studies (mainly of proteins), and finally groups focused on metabolic and population simulations.
Although interest in the field has increased in the last few years and we expect that the number of researchers and teams will rise outstandingly in the next years, we observe some problems that could hinder the sustainable development of this area in Argentina.
One of the problems we already referred to is the lack of integration of research related to computational biology in Argentina. Possibly, the creation of a specific institution or a society devoted to this activity could consolidate different groups and their interests. This would encourage collaborations and optimize the sharing of resources such as computational capacity or bibliography. It is important to mention that other countries in the region, such as Brazil, Chile, and Mexico, have their own reference centers (for example, the Bioinformatics Laboratory at EMBRAPA from Brazil,
Other problems could be related to economic issues and the lack of financial support from public or private resources to this field. It is important to note that the average investment in research and development in Latin American countries is about 0.2% of the gross domestic product [
There is no specific policy to promote computational biology in Argentina, but there is a plan to promote biotechnology, including bioinformatics (L. Barañao, Head of Agencia Nacional de Promoción Científica y Tecnológica, personal communication) [
Not all the problems are economic. The demand for computer scientists to work in close relation with experimental groups has progressively increased in the last two years. The problem that arises here, related to academic issues, is that universities, with the exception of Universidad Nacional de Entre Ríos, do not realize the value of computational biology either for its intrinsic academic significance or as a source of jobs for new graduates. To remedy this, biologically oriented degrees should include more flexible curricula to allow for the existence of different degree orientations.
Summing up, we think that there is a lot of work ahead to achieve sustainable growth of computational biology in Argentina. The quality of the academic production of the groups described, the existence of innovative basic research lines, and the study of systems with regional interest indicate a promising future. Together they reveal that in Argentina the field of computational biology is only just beginning.
We would like to thank Pablo Lorenzano, Diego Golombek, Nicolas Palopoli, Claudio Valverde, Julián Echave, Maria Silvina Fornasari, and Daniel Ghiringhelli for their suggestions in the writing of this manuscript. Lino Barañao and Mario Latuada provided us with official information from Agencia Nacional de Promoción Científica y Tecnológica and CONICET, respectively. We would also like to thank all the research groups discussed here, for their willingness to answer our questions about their activities.
Consejo Nacional de Investigaciones Científicas y Técnicas
Centro Regional de Estudios Genómicos
Instituto Nacional de Tecnología Agropecuaria
structurally constrained protein evolutionary model