Name
Abstract | ||
|---|---|---|
•Traditional virus origin methods consume lots of time and labor, our approach provides a new research orientation for hosts tracing from the data mining and offers an intelligent workflow to predict the host of newly discovered virus.•By collecting the genome sequences and annotating the hosts information, we build a benchmark of viral reference sequences to help researchers discover and develop the fight against virus.•Considering the problems of few annotated data and class-imbalance in virus datasets, we utilize the advantages of transfer learning and ensemble learning to conduct a HTL model which yields superiority than most of traditional methods.•We utilize HTL to give the host prediction for newly discovered virus (e.g., COVID-19). Being compared with traditional virus origin methods such as sequence identity and phylogenetic tree, we demonstrate the reliability of our method. |
Year | DOI | Venue |
|---|---|---|
2021 | 10.1016/j.jbi.2021.103736 | Journal of Biomedical Informatics |
Keywords | DocType | Volume |
COVID-19,Machine learning,Transfer learning,Ensemble learning,Hosts prediction,Virus origins | Journal | 117 |
ISSN | Citations | PageRank |
1532-0464 | 1 | 0.34 |
References | Authors | |
0 | 8 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Yun Yang | 1 | 1 | 1.02 |
| Jing Guo | 2 | 1 | 0.34 |
| Pei Wang | 3 | 1 | 0.34 |
| Yaowei Wang | 4 | 134 | 29.62 |
| Minghao Yu | 5 | 1 | 0.34 |
| Xiang Wang | 6 | 1 | 0.34 |
| Po Yang | 7 | 64 | 12.75 |
| Liang Sun | 8 | 1 | 0.34 |