Rizzo, E. | bioRxiv Science

bioRxiv2023

Digenic variant interpretation with hypothesis-driven explainable AI

MotivationThe digenic inheritance hypothesis holds the potential to enhance diagnostic yield in rare diseases. Computational approaches capable of accurately interpreting and prioritizing digenic combinations based on the probands phenotypic profiles and familial information can provide valuable assistance to clinicians during the diagnostic process. ResultsWe have developed diVas, a hypothesis-driven machine learning approach that can effectively interpret genomic variants across different gene pairs. DiVas demonstrates strong performance both in classifying and prioritizing causative pairs, consistently placing them within the top positions across 11 real cases (achieving 73% sensitivity and a median ranking of 3). Additionally, diVas exploits Explainable Artificial Intelligence (XAI) to dissect the digenic disease mechanism for predicted positive pairs. Availability and ImplementationPrediction results of the diVas method on a high-confidence, comprehensive, manually curated dataset of known digenic combinations are available at oliver.engenome.com.

bioinformatics↗

bioRxiv2017

Comparative Genomics Shows That Viral Integrations Are Abundant And Express piRNAs In The Arboviral Vectors Aedes aegypti And Aedes albopictus

BackgroundArthropod-borne viruses (arboviruses) transmitted by mosquito vectors cause many important emerging or resurging infectious diseases in humans including dengue, chikungunya and Zika. Understanding the co-evolutionary processes among viruses and vectors is essential for the development of novel transmission-blocking strategies. Arboviruses form episomal viral DNA fragments upon infection of mosquito cells and adults. Additionally, sequences from insect-specific viruses and arboviruses have been found integrated into mosquito genomes.\n\nResultsWe used a bioinformatic approach to analyze the presence, abundance, distribution, and transcriptional activity of integrations from 425 non-retroviral viruses, including 133 arboviruses, across the presently available 22 mosquito genome sequences. Large differences in abundance and types of viral integrations were observed in mosquito species from the same region. Viral integrations are unexpectedly abundant in the arboviral vector species Aedes aegypti and Ae. albopictus, but are [~]10-fold less abundant in all other mosquitoes analysed. Additionally, viral integrations are enriched in piRNA clusters of both the Ae. aegypti and Ae. albopictus genomes and, accordingly, they express piRNAs, but not siRNAs.\n\nConclusionsDifferences in number of viral integrations in the genomes of mosquito species from the same geographic area support the conclusion that integrations of viral sequences is not dependent on viral exposure, but that lineage-specific interactions exits. Viral integrations are abundant in Ae. aegypti and Ae. albopictus, and represent a thus far unappreciated component of their genomes. Additionally, the genome locations of viral integrations and their production of piRNAs indicate a functional link between viral integrations and the piRNA pathway. These results greatly expand the breadth and complexity of small RNA-mediated regulation and suggest a role for viral integrations in antiviral defense in these two mosquito species.

genomics↗