The bacteria talk

The bacteria talk

A reverse engineering approach for reconstructing the language of genes

“Let’s start from the end. Our project will not really end in our lifetime”. Puzzling as it may sound, the statement comes from Sarah Goldberg, researcher at the Technion – Israel Institute of Technology in Haifa, one of the leading scientific institutions in Israel. “Our objective is to uncover, at least partially, how functional elements of DNA interact with each other,” she tells

The impact of deciphering the communication protocols at the gene level could be huge. “If we were able to understand what goes wrong and why, for instance with cells, by capturing and then mutating the rules that instruct them to react in a given way, this would be an incredible step ahead for medicine”, Goldberg points out.

“At the moment I think we are quite advanced with bacteria, which are the main focus of our work at Technion. We are investigating the Escherichia coli and are developing a significant number of what we call the mutation libraries of it. You can think of mutation libraries as sets of rules, which contain information on what happens when functional elements in specific region interact.  Bacteria are actually relatively easy to investigate, because of their structure. As they are not as compact, their DNA is more accessible than, say, in a zebrafish. The idea with the libraries is that we can isolate data on a given region and then mutate that environment in a non-random way. By doing this, we can identify the parameters that affect gene expression, like its length, or what is binding to proteins and so on,” she adds. 

Goldberg is part of a multidisciplinary team of research organisations from all over Europe, busy working for almost two years on a project run under the FET (Future Emerging Technologies) programme of the European Union, with the title of MRG (Massive Reverse Genomics) Grammar.

In a way, we are trying to reconstruct the “grammar” rules by which gene expression is regulated, and determine the parameters influencing that expression. It’s a bit like reverse engineering a fairly complicated language. This is potentially an open-ended challenge. It will probably take years, a decade at least, to reach the stage when we could run clinical trials. In this sense, I don’t think there will be an end to our project”, Goldberg says.

Such exercises involve managing complex datasets and require the computational capacity to process them. However, this isn’t what worries researchers.  “Our bottleneck is not really how fast we can process data, but rather devising the strategies to design libraries and formulate the right algorithms, to extract the correlations of data that are relevant to what we do”, Goldberg concludes.

Algorithms are essential to reducing the number of variants that need to be examined. If that happens, the use of already existing genome editing tools will also be made more efficient. And the overall quest to understand the grammar of genes may end a bit earlier, if indeed there is an end.

Sarah Goldberg recently collaborated with the British artist Anna Dumitriu, who used the revolutionary technique CRISPR to edit the genome of an E. coli bacterium for the work “Make Do and Mend, inspired by the 75th anniversary of the first use of penicillin. The collaboration took place under the Future and Emerging Art and Technologies (FEAT) project.

For further information on Call +39 02 725 46 305. Email us at Or visit our Facebook, YouTube or Twitter.