Monday, November 21, 2016

How Big Data and Antibiotics Might Save You One Day



On October 31st, an international group of scientists released the details of a new method called DEREPLICATOR through an issue of Nature Chemical Biology. This method allows scientists for the first time to be able to find antibiotics that are hidden within mass spectrometry datasets.  Mass spectrometry is a process that allows researchers and scientists to identify different chemical structures of substances by separating the substance’s ions by mass and charge. The scientists are then able to run mass spectrometry data against a database of known antibiotics in order to detect known compounds in substances that had never before been analyzed. According to the article, roughly 2 million people within the United States develop an antibiotic resistance that renders certain antibiotics useless in combatting diseases. Never before, have scientists used “Big Data to look into microbial chemistry and characterize antibiotics and other drug candidates.” The way that DEREPLICATOR works is through an algorithm that searches through mass-spectrometry data to uncover “peptidic natural products,” (PNPs) which are a bioactive compound that have been known to include many antibiotics.

This new method, DEREPLICATOR, is a major breakthrough for researchers and scientists in the fight against antibiotic resistance. With the new algorithm researchers will also be able to discover new variants of known antibiotics. PNPs are apparently difficult to find due to their complex structures made of hundreds of non-standard amino acids as opposed to the standard 20 amino acids. Standard peptide identification tools, like SEQUEST, are therefore useless. SEQUEST is an analysis program that identifies collections of peptide sequences that have been generated from databases of protein sequences. As a result researchers will need to design new algorithms for antibiotic discovery that will be able to function in a “high-throughput technology” instead of just an academic setting. The goal of DEREPLICATOR is to do just that. To be able to handle high throughput would allow researchers to actually apply their new discoveries and variants of antibiotics.

Antibiotics are the first and most valuable line of defense against bacterial infections. Currently, if a person takes an antibiotic to fight a disease or infection for an extended period of time that antibiotic can eventually become ineffective in treating the disease or bacteria affecting the individual. Antibiotic resistance occurs when bacteria change in some way that reduces or eliminates the effectiveness of drugs, chemicals, or other agents designed to cure or prevent infections.This has happened due to doctors prescribing antibiotics to treat viruses that don’t respond to antibiotics to begin with causing them to lose their affect. The bacteria are then able to survive and continue to multiply over time, causing more harm by spreading the new form of the disease or bacteria that is resistant to antibiotics amongst other people. This is actually a rather uneasy situation that could potentially affect everyone one day. Without the use of antibiotics, we would be defenseless against bacterial diseases that have evolved to be immune to the affects antibiotics. DEREPLICATOR, however, may be able to stop this.

Source(s):



1 comment:

  1. I think David has brought up some excellent points in his argument. Though the DEREPLICATOR is a major breakthrough, I want to point out there are other methods besides the DEREPLICATOR. I wanted to expand on the potential problem of DEREPLICATOR may have, such as de-noising genetic information. In short, if we don’t de-noise the data, we would be presented with information that may be part of the problem or patter. However, it’s just extraneous information.

    Let’s transition to an alternate approach. The method that may resolve this issue is known as ADAGE (Analysis using De-noising Autoencoders of Gene Expression) and it’s another way to fight against antibiotic resistance. It’s similar to the method David mentioned, but ADAGE uses a de-noising technique in machine learning that identifies patterns in dirty data. ADAGE is able to identify genes that may be related and potentially “enhance and inhibit each others’ effects”.

    So how does this exactly help combat antibiotic resistance? ADAGE doesn’t need as much insight as DEREPLICATOR because (1) it’s a learning machine technique that may have or will “identified those genes as part of the problem” and (2) ADAGE can identify patterns of gene expression more efficiently and effectively. If you want to know more why we may need ADAGE, below is a video link genes identified in a malaria parasite. In this YouTube link, there are challenges in identifying patterns, which is virtually impossible
    https://www.youtube.com/watch?v=VwrF7lPvZc8



    https://www.extremetech.com/extreme/221812-machine-learning-offers-hope-in-fight-against-antibiotic-resistance

    ReplyDelete

Note: Only a member of this blog may post a comment.