Dr. Miklós Krész, Head of research department in Information processing at the InnoRenew CoE, together with colleagues Zoltán Maróti, Peter Juma Ochieng, József Dombi and Tibor Kalmár from the University of Szeged, published a scientific article “Optimizing sequence data analysis using convolution neural network for the prediction of CNV bait positions,” in BMC Bioinformatics.

Authors investigated the use of 1D convolutional neural network for accurately predicting the positions of oligonucleotide baits in complex whole-exome sequencing (WES) kits. Precisely determining these positions is crucial for effective data normalization and reducing systematic biases, which in turn improves the detection of copy number variations (CNVs). The study demonstrated that incorporating experimental coverage data, target region information, and sequencing data is essential for accurate predictions. This approach enhances CNV detection, which is important for genomic studies and may lead to a better understanding of the genetic basis of various diseases.