Biology subjects

Seno, S.

Publications and source records attributed to Seno, S..

2 recordsLinked to original sources

bioRxiv2020

A model ecosystem of twelve cryopreservable microbial species allowing for a non-invasive approach

Simultaneous understanding of both individual and ecosystem dynamics is crucial in an era marked by the degradation of ecosystem services. Herein, we present a high-throughput synthetic microcosm system comprising 12 functionally and phylogenetically diverse microbial species. These species are axenically culturable, cryopreservable, and can be measured noninvasively via microscopy, aided by machine learning. This system includes prokaryotic and eukaryotic producers and decomposers, and eukaryotic consumers to ensure functional redundancy. Our model system displayed both positive and negative interspecific interactions and higher-order interactions that surpassed the scope of any two-species interaction. Although complete species coexistence was not our primary objective, we identified several conditions under which at least one species from the producers, consumers, and decomposers groups, and one functionally redundant species, persisted for over six months. These conditions set the stage for detailed investigations in the future. Given its designability and experimental replicability, our model ecosystem offers a promising platform for deeper insights into both individual and ecosystem dynamics, including evolution and species interactions.

microbiology↗

bioRxiv2020

SC-JNMF: Single-cell clustering integrating multiple quantification methods based on joint non-negative matrix factorization

AO_SCPLOWBSTRACTC_SCPLOWUnsupervised cell clustering is important in discovering cell diversity and subpopulations. Single-cell clustering using gene expression profiles is known to show different results depending on the method of expression quantification; nevertheless, most single-cell clustering methods do not consider the method. In this article, we propose a robust and highly accurate clustering method using joint non-negative matrix factorization (joint NMF) based on multiple gene expression profiles quantified using different methods. Matrix factorization is an excellent method for dimension reduction and feature extraction of data. In particular, NMF approximates the data matrix as the product of two matrices in which all factors are non-negative. Our joint NMF can extract common factors among multiple gene expression profiles by applying each NMF to them under the constraint that one of the factorized matrices is shared among the multiple NMFs. The joint NMF determines more robust and accurate cell clustering results by leveraging multiple quantification methods compared to the conventional clustering methods, which uses only a single quantification method. In conclusion, our study showed that our clustering method using multiple gene expression profiles is more accurate than other popular methods.

bioinformatics↗