“It is possible to simulate life with a program, including all the wonderful, mysterious and incredibly complex biochemistry that makes it work?”
This is the question che Markus Covert, assistant professor of bioengineering, and his team have set themselves at Stanford University.
This would make it possible to test experimental ideas prior to spend time and money to do it in the laboratory.
A project, this, which involves engineers for many years, encountering, however, insurmountable obstacles that led to the failure of every attempt.
Illumination comes in 2008. Creating a simulator simple and practical: choosing one of the simplest single-celled existing microorganisms, a bacterium called Mycoplasma Genitalium, and build a model of one of these germs.
The winning idea was, therefore, to simulate a single cell -not a cell system- thus including any known biological event: the opening of the double helix of DNA, transcription of messages in a DNA copy of RNA, the production of each enzyme and protein based on the instructions contained in the RNA and all other processes that enable cell growth and division.
In seeking to achieve this goal, the team was inspired by the researcher who first dreamed of modeling life: Harold Morowitz who, twenty-four years before, had thought that simplest bacteria, mycoplasmas, was the right starting point.
“Each experiment that can be performed in the laboratory can also be made to the computer. The extent to which the experimental results and those of the simulation match is a measure of the completeness of the biology paradigm”, Morowitz claimed.
Before they could simulate the life cycle of a microbial species, they had to solve three problems. First, it was necessary to encode in mathematical formulas and computer algorithms all relevant functions, from the energy flows to the synthesis and decomposition of DNA, RNA and proteins.
Second, they needed a general platform to integrate all these functions. Last problem: determining the upper and lower limits for each of the approximately 1700 model parameters in order to obtain biologically realistic values.
There winning idea plays its role: for a single cell division is a dramatic event. Prior to be divided the organism must double its mass as well as the amount of DNA. If the model is limited to a single cell, the computer can really count and follow each molecule during the entire cell cycle.
The team set up by Markus Covert consists of physicists, biologists, modelers, and also a software engineer who worked at Google. Among the engineers there are two bioengineers: Michael Schuler pioneer of the simulation cell, and Bernhard Palsson, creator of flux balance analysis technique for metabolism.Source: cdn.theatlantic.com
The cell model created is composed of 28 distinct modules, each of which uses its own algorithm. This model is based on the operation of HYSYS simulation package, used by Covert during his university studies for the project of a refinery.
Thus, as these processes take place simultaneously in a cell, their actions are independent of periods of less than one second. So the cell was divided in steps of one second, running the modules in order.
During the early stages, each simulation produced 500 megabytes of data: numerical results arrived at a control panel through dozens of charts that correspond to stacks of printed sheets.
First results were quite far from reality. Day by day we made changes to the code and mathematics became more sophisticated until the day when the results were considered plausible.
Many successes have been achieved through this approach: it has been observed that proteins are expelled from the DNA about 30 thousand times over nine hours in the life cycle; it could explain why the cell stops dividing immediately when certain genes are deactivated but reproduces ten more times before dying when other essential genes are turned off.
For now the finish line saw the simulation of a cell of Mycoplasma Genitalium and was the first step towards a computer modeling of human cells and tissues at the level of genes and molecules. Compared to bacteria, human cells have many more subdivisions and show a much greater genetic control, which largely is not known anything yet.
The team has published the code of the software giving everyone the opportunity to run it on their machines. Close-up Engineering has decided to show it to readers by inserting the software in the pages of the site. By clicking on the links below, you can see the different structures implemented in the program (after you have clicked on the link, wait 5 seconds and proceed by clicking “Next” in the upper right).
COMPUTATIONAL SIMULATION OF A MYCOPLASMA GENITALIUM CELL – CYCLE AND 3D SHAPE
COMPUTATIONAL SIMULATION OF A MYCOPLASMA GENITALIUM CELL – DNA REPLICATION
COMPUTATIONAL SIMULATION OF A MYCOPLASMA GENITALIUM CELL – SYNTHESIS
COMPUTATIONAL SIMULATION OF A MYCOPLASMA GENITALIUM CELL – POPULATION
- Lee R, Karr JR, Covert MW. WholeCellViz: Data Visualization for Whole-Cell Models. BMC Bioinformatics 14, 253 (2013). BMC Bioinformatics | PubMed | SimTK
- Karr JR et al. A Whole-Cell Computational Model Predicts Phenotype from Genotype. Cell 150 2, 389-401 (2012). Cell |PubMed | SimTK
(Source header image: http: hu.wallpapersus.com)
Close-up Engineering – ©allrightsreserved
Founder di Close-up Engineering e CEO di Vibre, azienda che si occupa di interfacce neurali.
Nel 2020 è stato inserito nella lista di Forbes Italia tra i 100 giovani under 30 più influenti.