What makes a human different from a chimp?  Researchers from the European Molecular Biology Laboratory’s European Bioinformatics Institute [EMBL-EBI] have come one important step closer to answering such evolutionary questions correctly.  In the current issue of Science they uncover systematic errors in existing methods that compare genetic sequences of different species to learn about their evolutionary relationships.  They present a new computational tool that avoids these errors and provides accurate insights into the evolution of DNA and protein sequences.  The results challenge our understanding of how evolution happens and suggest that sequence turnover is much more common than assumed.

“Evolution is happening so slowly that we cannot study it by simply watching it.  That’s why we learn about the relationships between species and the course and mechanism of evolution by comparing genetic sequences,” says Nick Goldman, group leader at EMBL-EBI.

The four letter code that constitutes the DNA of all living things changes over time; for example individual or several letters can be copied incorrectly [substitution], lost [deletion] or gained [insertion].  Such changes can lead to functional and structural changes in genes and proteins and ultimately to the formation of new species.  Reconstructing the history of these mutation events reveals the course of evolution.

A comparison of multiple sequences starts with their alignment.  Characters in different sequences that share common ancestry are matched and gains and losses of characters are marked as gaps.  Since this procedure is computationally heavy, multiple alignments are often built progressively from several pairwise alignments.  It is impossible, however, to judge if a length difference between two sequences is a deletion in one or an insertion in the other sequence.  For correct alignment of multiple sequences, distinguishing between these two events is crucial.  Existing methods, that fail to do that, lead to a flawed understanding of the course of evolution.

“Our new method gets around these errors by taking into account what we already know about evolutionary relationships,” says Ari Löytynoja, who developed the tool in Goldman’s lab.  “Say we are comparing the DNA of human and chimp and can’t tell if a deletion or an insertion happened.  To solve this our tool automatically invokes information about the corresponding sequences in closely related species, such as gorilla or macaque.  If they show the same gap as the chimp, this suggests an insertion in humans.”

Findings achieved with the new technique suggest that insertions are much more common than assumed, while the frequency of deletions has been overestimated by existing methods.  A likely reason for these systematic errors of other techniques is that they were originally developed for structural matching of protein sequences.  The focus of molecular biology is shifting, however, and understanding functional changes in genomes requires specifically designed methods that consider sequences’ histories.  Such approaches will likely reveal further bugs in our understanding of evolution in future and might challenge the conventional picture of sequence evolution.

Two papers published in the journal Science today* by Microsoft Research ecologist Drew Purves together with research colleagues at Princeton University and universities in Madrid, Spain, highlight how an improved understanding of forest dynamics is needed to better predict environmental change. The research suggests that a new generation of realistic forest modelling, which is urgently needed and now within reach, will significantly improve an understanding of how forests work, how tree species respond to deforestation, and how forests impact climate regulation and environmental change.

The research points out that forest dynamics (how populations of trees interact with each other and the environment) remains the single most important outstanding component in fully understanding climate change. There trillions of trees on the planet, made up of more than 100,000 species, which contain as much carbon as is currently in the atmosphere and serve as home to two-thirds of the planet’s terrestrial biodiversity. However, while other climate change factors such as ocean dynamics are now well researched, the effects of changes to the world’s forests are still largely unknown.

The paper “Predictive Models of Forest Dynamics” by Purves and Princeton’s Stephen Pacala explores dynamic global vegetation models (DGVMs), which simulate the reaction of forests to past, present and future climate.

“DVGMs have shown that forests could be a crucial part of the way the Earth’s climate responds to man-made CO2 emissions, but insufficient understanding of forests, and insufficient data and computing power, have made their predictions highly uncertain,” Purves said. “This kind of uncertainty helps climate sceptics, who erroneously conclude that because the Earth is a complex but poorly understood system, we should not change our behaviour. However, we suggest that the convergence of recently developed mathematical models, improved data sources and new methods in computational data analysis could produce more realistic models. That would give us truly invaluable information to help manage the world’s forests and understand their impact on our climate.”

“Until now, one of the most important pieces of the climate change jigsaw has been missing,” Pacala said. “We argue that we can significantly further our understanding of forest dynamics if scientists work together to use new computational techniques and data sources — provided governments and others make more data available in useful forms. We feel that these discoveries could unlock the climate change mysteries of forests on a global scale in as little as five years.”

The second paper published in Science today, “Animal vs Wind Dispersal and the Robustness of Tree Species to Deforestation,” by Daniel Montoya from the Universidad de Alcalá in Madrid and Purves in Cambridge, with Miguel A. Rodríguez of the Universidad de Alcalá and Miguel A. Zavala of Centro de Investigación Forestal, Instituto Nacional de Investigación y Tecnología Agraria y Alimentaria (INIA-CIFOR) in Madrid, examines what happens to individual tree species in the face of deforestation. Using data from nearly 90,000 survey plots in the Spanish peninsula, the paper found tree species that rely on wind to disperse seeds, rather than animals, are more vulnerable to deforestation.

Montoya said, “By applying various methods in computational data analysis to a large source of forest data, we have confirmed that, in Spain at least, plants with animal-dispersed seeds are less vulnerable to habitat loss, because animals provide trees with an intelligent dispersal mechanism, travelling and distributing seeds between areas of remaining forest. In contrast, a wind dispersal method is more susceptible to habitat loss, as seeds are more likely to fall in inhospitable environments. Using methods like this, conservationists can identify the species at most risk following deforestation, and use this knowledge to develop new strategies to mitigate the effects of widespread habitat loss and help to protect species diversity.”

The research also concludes that when no animal dispersers exist in the ecosystem, animal-dispersed tree species are the most vulnerable to deforestation. This means that protecting plant-animal interactions must also be a cornerstone of conservation policy, because the interactions not only create and maintain biodiversity, but also increase resistance to disturbances to the ecosystem.

Both papers underline the importance of forest dynamics in understanding and predicting climate change and biodiversity, highlighting the urgent need for additional study and resources. Purves said, “It is imperative that we create the tools and science to accurately understand the reaction of ecosystems to climate change and other forces — not just for plants and animals, but for our children and succeeding generations.”

This research is part of the recently established Computational Science Research at Microsoft Research Cambridge. This team of ecologists, biologists, neuroscientists, mathematicians and computer scientists is pioneering novel theoretical frameworks, computational tools and scientific methods to tackle the greatest scientific and societal challenges of this century, from climate change and declining biodiversity to understanding how living things work.

US researchers have created ‘living computers’ by genetically altering bacteria. The findings of the research, published in BioMed Central’s open access Journal of Biological Engineering, demonstrate that computing in living cells is feasible, opening the door to a number of applications including data storage and as a tool for manipulating genes for genetic engineering.

A research team from the biology and the mathematics departments of Davidson College, North Carolina and Missouri Western State University, Missouri, USA added genes to Escherichia coli bacteria, creating bacterial computers able to solve a classic mathematical puzzle, known as the burnt pancake problem.

The burnt pancake problem involves a stack of pancakes of different sizes, each of which has a golden and a burnt side. The aim is to sort the stack so the largest pancake is on the bottom and all pancakes are golden side up. Each flip reverses the order and the orientation (i.e. which side of the pancake is facing up) of one or several consecutive pancakes. The aim is to stack them properly in the fewest number of flips.

In this experiment, the researchers used fragments of DNA as the pancakes. They added genes from a different type of bacterium to enable the E. coli to flip the DNA ‘pancakes’. They also included a gene that made the bacteria resistant to an antibiotic, but only when the DNA fragments had been flipped into the correct order. The time required to reach the mathematical solution in the bugs reflects the minimum number of flips needed to solve the burnt pancake problem.

“The system offers several potential advantages over conventional computers” says lead researcher, Karmella Haynes. “A single flask can hold billions of bacteria, each of which could potentially contain several copies of the DNA used for computing. These ‘bacterial computers’ could act in parallel with each other, meaning that solutions could potentially be reached quicker than with conventional computers, using less space and at a lower cost.” In addition to parallelism, bacterial computing also has the potential to utilize repair mechanisms and, of course, can evolve after repeated use.