ePoster

Machine learning-based exploration of long noncoding RNAs linked to perivascular lesions in the brain

Hiyori Edo, Ryodai Itano, Masakazu Umezawa
FENS Forum 2024(2024)
Messe Wien Exhibition & Congress Center, Vienna, Austria

Conference

FENS Forum 2024

Messe Wien Exhibition & Congress Center, Vienna, Austria

Resources

Authors & Affiliations

Hiyori Edo, Ryodai Itano, Masakazu Umezawa

Abstract

Perivascular lesions with accumulation of proteins with abnormal secondary structures are associated with neurodegenerative diseases such as dementia. Along with upregulation of the genes (Gfap, Aqp4) involved in the pathogenesis, a number of long noncoding RNAs are differentially expressed in the lesion; however, the roles and functions of such noncoding RNAs are unknown. This study was aimed to explore novel functions of long noncoding RNAs (lincRNAs) potentially involved in the suppression of the perivascular lesions. RNA expression profiles in the cerebral cortex of mice (6 weeks old) with perivascular lesion model (dataset: GSE250286) were analyzed using principle component analysis to extract lincRNAs highly correlated to the lesion level. After extraction of common sequences of the lincRNA groups and mRNAs having their complementary sequences, the mRNAs extracted and expressed in the cerebral cortex were subjected to functional analysis by DAVID. We found a common sequence in the lincRNAs inversely correlated with the lesion extent. The functions of mRNAs with the complementary sequences were enriched in 8 Gene Ontology, including “cell projection”, “dendrites”, “synapses”, and “projection neurons”, which play roles in the elongation of astrocytic endfoot observed in the perivascular lesions, and in neural functions in the brain. The lincRNAs and their common sequences extracted in this study can potentially regulate cellular development of the lesions. The effect of RNA molecules with this sequence on the control of the pathogenesis, which can be linked to neurodegenerative and neurodevelopmental disorders, is of future interest.

Unique ID: fens-24/machine-learning-based-exploration-31b663c4