| Autor: |
Yang, Xinan H, Wang, Zhezhen, Feliciano, Antonio, Griggs, Dannie, Tang, Fangming, An, Qier, Cunningham, John M |
| Jazyk: |
angličtina |
| Rok vydání: |
2019 |
| DOI: |
10.1101/668442 |
| Popis: |
Summary Abrupt and irreversible changes (or tipping points) are decisive in disease and normal phenotypic progress. Often, however, computational approaches detect critical-transition signals (CTSs) indicating tipping points from longitudinal data – which often are not available for patient transcriptomes. Here we adopt historical tipping-point approaches to cross-sectional data by modeling high probability spaces of phenotypes. We formulate this task as a generalized CTS-searching problem and derive a robust algorithm to solve it. We construct a comprehensive scoring scheme and successfully apply the scheme to lymphoma, lung-injury, heart-development, and neuroblastoma systems. Thus, we identify a spatial gene-expression feature for systematic dynamics at phenotypic tipping points, which can be exploited to infer functional genetic variations and transcription factors. Our framework (‘BioTIP’) can analyze not only time-course but also cross-sectional transcriptomes and is compatible with noncoding RNA profiles. Additional knowledge discovery that explores the critical transition of a system can be tested using our approach. Highlights Adopting tipping-point theory to transcriptomes Robust framework for the identification of critical-transition signal at tipping points Statistics on a high probability space enables the tipping-point analysis of cross-sectional data Application to gene expression of neuroblastoma reveals a gene regulatory network transition and underlying mechanisms |
| Databáze: |
OpenAIRE |
| Externí odkaz: |
|
