GWIscan: an automated pipeline for genome-wide identification and in silico characterization of gene families
GWIscan is an automated, multi-species pipeline that integrates BLAST, profile HMM, and InterPro analyses to comprehensively identify and characterize gene families, outperforming single-method approaches by recovering additional high-confidence members and reproducing established inventories.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer
Life is built from instructions written in a code called DNA, which cells read to build proteins. These proteins are the workers that carry out nearly every task inside an organism, from digesting food to defending against disease. Often, these workers come in groups, where many members share a similar shape or function because they descended from a common ancestor. Scientists call these groups gene families. To understand how a living thing works, researchers must first find and count these families within its genetic code. However, as new genomes are sequenced at a rapid pace, the sheer volume of data has made this task difficult. Traditional methods rely on searching for matches between known protein shapes and new genetic sequences, but no single search method is perfect. Some families are missed because they look too different from the known examples, while others are falsely identified because they share a small, unrelated feature. This leaves researchers with incomplete lists and the tedious job of manually checking and organizing the results.
A new tool called GWIscan, developed by Beaven Manjengwa, offers a solution by automating this entire process. The software acts as a comprehensive scanner that looks for gene families across an entire genome, combining two different search strategies to ensure nothing is missed. It first looks for direct matches to known protein sequences, similar to finding a specific book by its title. It then performs a second, more subtle search that looks for the underlying structural blueprint of the protein, even if the title has changed. By merging the results of these two searches, the tool creates a more complete list of candidates. It then acts as a strict editor, verifying that each candidate truly belongs to the family by checking for specific, defining features. Once the list is confirmed, the software automatically gathers detailed information about every member, such as its size, chemical properties, and where it is likely to work inside the cell. This entire workflow happens in a single, automated run, removing the need for researchers to stitch together different programs or manually compile data.
To test if this approach worked, the researchers applied GWIscan to the genetic codes of three different plants: a small flowering weed, corn, and a common bean. They focused on a specific group of proteins called lectins, which help plants recognize and interact with their environment. The team compared the results from GWIscan against previous studies that had manually counted these proteins in the same plants. In the weed, the new tool found 211 of the 217 genes that had been reported in earlier research, a success rate that matched or improved upon previous efforts. In the corn and bean studies, the tool not only reproduced the known counts but also found additional members that the older, single-method searches had overlooked. For instance, in the bean plant, the tool identified dozens of new lectin genes that were missed because they had a slightly different structure than the standard examples. These new findings were not random guesses; the tool confirmed that these extra genes possessed the correct structural features to be part of the family.
The study demonstrates that combining different search methods is more effective than relying on just one. When the researchers looked closely at the genes that were missed by older methods, they found that some were too different in sequence to be caught by a simple match, while others were too complex for a structural search alone. By using both approaches together, GWIscan recovered these missing pieces. The tool also proved capable of handling data from multiple species at once, organizing the results into a clear comparison that shows how gene families vary across different plants. While the software is designed to work on any living organism, the researchers noted that some of the advanced analysis steps require specific licenses for academic use. Ultimately, GWIscan provides a reliable, automated way for scientists to map the genetic families of any species, turning a complex, manual task into a streamlined process that yields high-confidence results.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.