New bioinformatics tools have been developed to help identify harmful genetic changes in the human genome, according to two studies published in the journals NAR Genomics & Bioinformatics and Genome Biology. Researchers from Martin Vingron's laboratory at the Max Planck Institute for Molecular Genetics created these tools to improve the diagnosis of genetic diseases by better analyzing complex genetic differences. These differences, known as structural variants and tandem repeats, can affect large sections of DNA and are increasingly linked to various health conditions. However, analyzing them has been challenging for scientists due to their complexity. The most common method for genetic testing, called short-read sequencing, reads only short segments of DNA. This makes it difficult to detect structural variants, which are often larger than the DNA segments that can be read. To overcome this, the research team used machine learning to recognize patterns in known structural variants, allowing them to classify new variants more accurately. This approach reduces errors, which were common in older methods that required manual review. A special type of genetic change, called tandem repeats, occurs when short DNA sequences are repeated many times in a row. These repeats vary widely between individuals and are linked to both healthy and diseased states. They are important in forensic science and paternity testing, and they are also associated with several neurological diseases. The researchers developed a tool named "Dicast" that successfully identified all harmful structural variants while filtering out many false positives. They also created an algorithm that can precisely count the number of these repeated sequences from sequencing data. In the second study, the team introduced an algorithm and a visualization tool that help researchers quickly identify harmful tandem repeats in patient data. Led by first authors Lion Ward Al Raei and Maryam Ghareghani, the researchers tested their methods using real-world examples, showing their effectiveness. These tools are already being used in research and could soon be integrated into diagnostic processes to improve the accuracy of rare disease diagnoses. Some researchers are working to further explore the potential of these genetic changes through a startup called Lucid Genomics. Advances in sequencing technology, paired with these new analysis tools, are expected to significantly enhance genetic diagnostics in the future.