Researchers from Tel Aviv University have shown that silent mutations, i.e. those that do not change the amino acid sequence of the proteins, are also not innocent
Researchers from the Department of Biomedical Engineering and the Zimin Institute for World-changing Engineering Research at Tel Aviv University were able to predict the type of cancer and the death rate from it according to "silent" mutations in the cancer genomes - proof of the feasibility that could save lives in the future. The results of the research, led by Prof. Tamir Toler and research student Tal Gutman, are published today in the journal NPJ Genomic Medicine.
"Silent" mutations are defined as those that do not change the amino acid sequence of the proteins. In recent years, evidence is accumulating that silent mutations, both inside and outside the genetic coding region in the cell, can affect gene expression and that they may be related to the development and spread of cancer cells. However, so far it has not been quantitatively tested whether these mutations can contribute to identifying the type of cancer and predicting the patient's chances of survival.
In the new study, based on approximately three million mutations in the cancer genomes of 9,915 patients, the researchers tested whether they could identify the type of cancer and estimate the death rate from it 10 years after the initial diagnosis based solely on the silent mutations - and found that the prediction ability of the silent mutations is much similar Cases for the performance of the conventional predication of the "normal" mutations. In addition, the researchers tried to evaluate whether a combination of information on silent and normal mutations can improve the ability to classify the type of cancer, and found that the information obtained from silent mutations improves the margin of error by 68%. In certain types of cancer, this is an improvement of up to 17% in the ability to classify, while combining the two types of mutation can improve the prognosis by up to 5%.
Mutations involved in turning a healthy cell into a cancerous cell
"In our genome, as in every genome of any other living creature, there are mutations that can change the amino acid sequence of the proteins encoded in the genome," explains Prof. Toler. "Since these proteins are responsible for the various mechanisms in the cell, such mutations are involved in turning the healthy cell into a cancerous cell. In contrast, there are mutations that do not change the amino acids, so they were called 'silent' and were ignored for many years. For the first time, we conducted analyzes of approximately 10,000 cancer genomes of all types, and showed that the silent mutations have a diagnostic value, which type of cancer it is, and also a prognostic value, how long the patient will survive."
According to Prof. Toler, the genetic material in the cell holds two types of information: what is the amino acid sequence that is produced - and when and how much to produce of each protein - that is, the regulation of the production process. "Those silent mutations can affect the regulation of gene expression, and this is an effect no less important than the type of protein that is produced. Obviously if the cell produces much less of a certain protein - it is almost as bad as deleting it. Another effect is the folding of the protein. The protein is a long molecule that usually includes many hundreds of amino acids, where the three-dimensional folding of the molecule begins as soon as they are produced in the ribosome. The production rate of the protein by the ribosome affects the folding, and the silent mutations can affect the production rate of the protein and therefore its folding - a folding that is significant for actual function. In addition, there are cases where the silent mutations affect a process called splicing, where pieces of the genetic material are cut to create the final sequence from which the protein will be formed. In short, it turns out that these silent mutations make a lot of noise, and we were able to quantify their impact for the first time."
To test their hypothesis and quantify the effect of these mutations, Prof. Toler and his colleagues used public genetic information on cancer genomes from the National Institutes of Health (NIH) in the US. The researchers took the data on the cancer genome and tried using methods based on machine learning to predict the type of cancer and how many years each patient lived according to the silent mutations - and then compared the results they received to the true data from the database.
"The results of the study have several important implications," says Prof. Toler. "First of all, using silent mutations can definitely improve models that predict prognosis and are used for classification. It is important to note that even a 17% improvement has a great meaning because behind these numbers are people we love, and one day we may be ourselves, therefore every percentage improvement is dramatic. A doctor who discovers metastases wants to know the source of the metastasis and the course of the disease's development, in order to adjust the best treatment. If, for example, instead of wrong diagnosis and prognosis for five out of ten cancer patients, we reach a situation where only four out of ten cancer patients are wrong, this could eventually translate into millions of patients whose lives could possibly be saved. In addition, our results show that only based on silent mutations it is possible in many cases to obtain similar predictive performance to relying on mutations that are not silent. This is an encouraging result, because in recent years technologies have been developed that classify cancer based on relatively non-invasive blood tests, based on the analysis of pieces of DNA from a cancerous source. Since most of our DNA does not code for protein, it is likely that most of the pieces of this type that we fish will contain silent mutations."
The new research has implications for all fields of oncology research and treatment, and after proving this feasibility, the researchers intend to establish a startup with a "Sanra" incubator, which will focus on silent mutations as a medical tool for all intents and purposes.
More of the topic in Hayadan:
