Simple Science

Cutting edge science explained simply

# Biology # Bioinformatics

Riboswitches: Tiny RNA Switches with Big Impact

Discover how riboswitches control protein production in cells and their potential health implications.

William S. Raymond, Jacob DeRoo, Brian Munsky

― 6 min read


Riboswitches: RNA's Riboswitches: RNA's Hidden Control protein regulation and disease. Riboswitches reveal RNA's role in
Table of Contents

A riboswitch is a special part of RNA that acts like a switch to control certain processes in a cell. Think of it like a light switch that can be turned on or off depending on whether a specific small molecule is present. When this small molecule binds to the riboswitch, it can change the shape of the RNA. This change can either reveal or hide parts of the RNA that are important for making Proteins. The riboswitch can thus control how much protein is made, which in turn affects various functions within the cell.

How Does a Riboswitch Work?

Riboswitches have two main shapes or forms. When the small molecule is not present, the riboswitch might be in one form that allows the cell to make a certain protein. But when the small molecule comes along and binds to the riboswitch, it changes to another shape. This new shape can stop protein production or allow it to continue, depending on the situation. It’s like switching between playing a game or taking a nap-what you do depends on the circumstances!

Where are Riboswitches Found?

Riboswitches are mostly found in bacteria (prokaryotes), where they can control around 40 different Small Molecules. However, in more complex organisms like plants and animals (eukaryotes), riboswitches are not as common. The few that have been found in plants usually respond to a molecule called thiamine pyrophosphate (TPP). In humans, they are less well-studied, leading scientists to ask if there are hidden riboswitches awaiting discovery.

Why Are Riboswitches Important?

Riboswitches can have big implications for health and disease. If a riboswitch is not working properly, it could lead to either too much or too little of a protein being made. This can contribute to diseases, either by producing a harmful amount of a protein or by not producing enough of a needed one. So understanding riboswitches could help scientists learn more about certain diseases and possibly find new treatments.

How Are Scientists Finding Riboswitches?

Research into riboswitches has been ongoing since they were discovered in 2002. Scientists have developed various methods to identify them, including computer programs that can predict where riboswitches may be located in RNA sequences. These programs often use advanced techniques, including Machine Learning, to analyze vast amounts of genetic information.

The Role of Machine Learning in Riboswitch Discovery

Recently, scientists have started using machine learning to help find riboswitches. Machine learning uses algorithms to learn from data and make predictions on new data. In this case, researchers trained computer models using known riboswitch sequences to help identify potential new riboswitches in human RNA sequences.

They gathered a large number of riboswitch sequences and human RNA sequences to create a model that could classify new sequences as likely or unlikely to be riboswitches. The model was then tested to see how well it could predict riboswitches it hadn’t seen before.

Building the Model

Scientists collected a massive pool of riboswitch sequences from various databases, cleaning and organizing the data to make it useful for training. They then extracted many features from the RNA sequences, such as their length, structure, and other characteristics.

Next, they created several machine learning classifiers-think of these as digital detectives trained to recognize riboswitches based on previous examples. They validated these classifiers by testing them on sequences that were carefully selected to ensure they had not been used in training.

The Results

After running their models, researchers found a number of sequences in human RNA that seemed promising as potential riboswitches. They identified 436 sequences that all classifiers agreed on, suggesting they were likely candidates for harboring riboswitch-like features.

Additionally, they noticed that a larger pool of 1,533 sequences also showed riboswitch properties, although these were identified by only a subset of classifiers. This insight gave scientists a solid starting point for future experimental investigations.

What Makes These Riboswitch Candidates Special?

The identified sequences shared many features with known riboswitches. Researchers took a closer look at these hits to assess their characteristics, and they even set up an online display to compare these potential riboswitches against known sequences. This approach not only helps in organizing the information but makes it easier for other researchers to explore these findings further.

Gene Ontology: What Do the Hits Mean?

Researchers also performed a Gene Ontology (GO) analysis on the 5’ UTR hits to understand what functions these potential riboswitches might impact. They found that many of the proteins related to the hits were involved in essential processes such as gene regulation, mRNA processing, and even detecting chemical signals. This suggests that these riboswitches could play significant roles in responding to small molecules within the cell.

Potential Implications for Health

The identification of these riboswitch-like sequences sets the stage for important future work. If any of the discovered sequences truly behave like riboswitches, they could provide insights into how cells regulate themselves under different conditions. For instance, riboswitches could be key players in diseases where the normal regulation of proteins is disrupted.

Conclusion

The study of riboswitches presents an exciting opportunity for scientists to learn how cells control various processes. The use of machine learning and computational tools has proven to be an innovative way to sift through vast amounts of genetic data, revealing new candidates for riboswitches. While there’s still much work to be done-like validating these findings through experiments-the technology and methods being used are paving the way for a deeper understanding of genetics and cellular function. Who would have thought that tiny nucleotides could hold such immense potential?

Next Steps in Riboswitch Research

As researchers continue to investigate riboswitches, they aim to conduct more experiments to validate the computational predictions. By studying these riboswitches, scientists hope to uncover their roles in various biological processes and diseases.

Furthermore, future research will likely extend to other areas of RNA biology, including the exploration of riboswitches in different organisms and how they can be manipulated for therapeutic purposes.

With ongoing advancements in technology and a growing appreciation for the complexity of RNA, the world of riboswitches is bound to offer even more surprises. Stay tuned; who knows what other secrets this tiny world of RNA has in store!

Original Source

Title: Identification of potential riboswitch elements in Homo Sapiens mRNA 5'UTR sequences using Positive-Unlabeled Machine learning

Abstract: Riboswitches are a class of noncoding RNA structures that interact with target ligands to cause a conformational change that can then execute some regulatory purpose within the cell. Riboswitches are ubiquitous and well characterized in bacteria and prokaryotes, with additional examples also being found in fungi, plants, and yeast. To date, no purely RNA-small molecule riboswitch has been discovered in Homo Sapiens. Several analogous riboswitch-like mechanisms have been described within the H. Sapiens translatome within the past decade, prompting the question: Is there a H. Sapiens riboswitch dependent on only small molecule ligands? In this work, we set out to train positive unlabeled machine learning classifiers on known riboswitch sequences and apply the classifiers to H. Sapiens mRNA 5UTR sequences found in the 5UTR database, UTRdb, in the hope of identifying a set of mRNAs to investigate for riboswitch functionality. 67,683 riboswitch sequences were obtained from RNAcentral and sorted for ligand type and used as positive examples and 48,031 5UTR sequences were used as unlabeled, unknown examples. Positive examples were sorted by ligand, and 20 positive-unlabeled classifiers were trained on sequence and secondary structure features while withholding one or two ligand classes. Cross validation was then performed on the withheld ligand sets to obtain a validation accuracy range of 75%-99%. The joint sets of 5UTRs identified as potential riboswitches by the 20 classifiers were then analyzed. 15333 sequences were identified as a riboswitch by one or more classifier(s) and 436 of the H. Sapiens 5UTRs were labeled as harboring potential riboswitch elements by all 20 classifiers. These 436 sequences were mapped back to the most similar riboswitches within the positive data and examined. An online database of identified and ranked 5UTRs, their features, and their most similar matches to known riboswitches, is provided to guide future experimental efforts to identify H. Sapiens riboswitches. Author summaryRiboswitches are an important regulatory element mostly found in bacteria that have not been described in Homo Sapiens. However, if human riboswitches exist and if they can be found, they could have vast implications on human disease. We apply positive-unlabeled machine learning to on known riboswitch sequences to search H. Sapiens 5UTR sequences for potential riboswitches. We analyze our ensemble predictions for likely H. Sapiens 5UTR riboswitches using GO analysis to determine their potential functional roles, and we rank and display our predicted sequences next to the most similar known riboswitches. We expect these analyses to be helpful to the scientific community in planning future experiments for laboratory discovery and validation. 0.1 Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=169 SRC="FIGDIR/small/568398v2_ufig1.gif" ALT="Figure 1"> View larger version (55K): [email protected]@10f085corg.highwire.dtl.DTLVardef@1edfcaborg.highwire.dtl.DTLVardef@1674ce0_HPS_FORMAT_FIGEXP M_FIG C_FIG

Authors: William S. Raymond, Jacob DeRoo, Brian Munsky

Last Update: 2024-12-06 00:00:00

Language: English

Source URL: https://www.biorxiv.org/content/10.1101/2023.11.23.568398

Source PDF: https://www.biorxiv.org/content/10.1101/2023.11.23.568398.full.pdf

Licence: https://creativecommons.org/licenses/by/4.0/

Changes: This summary was created with assistance from AI and may have inaccuracies. For accurate information, please refer to the original source documents linked here.

Thank you to biorxiv for use of its open access interoperability.

Similar Articles