biomvRCNS

biomvRCNS introduces a sophisticated hidden semi-Markov model (HSMM) designed to enhance genome annotation by facilitating the detection of transcripts, regulatory regions, and copy number variations from the vast datasets generated by high-throughput technologies such as tiling arrays and next-generation sequencing (NGS). These technologies produce continuous segments or signal peaks in the genome indicative of various biological features, including transcript variants, regions of deletion or amplification, and epigenetic modifications. The sheer volume and complexity of the data pose significant analytical challenges.

The HSMM implemented in biomvRCNS is tailored to handle multiple genomic profiles, offering a versatile approach to genome annotation across different applications. Unlike models confined to specific data types or genomic features, biomvRCNS supports various data distributions, making it a general segmentation engine suitable for a broad range of genomic data types. A vital part of the model is its incorporation of genomic positions into the sojourn distribution, enhancing the biological relevance of the modeling output. This aspect, coupled with the option to incorporate prior knowledge from annotations or previous studies, ensures that the results are statistically sound and biologically plausible.