Table 4.
Machine learning models based on attribute weighting models demonstrated that the developed transcriptomic signature of lactation is independent from the species.
| Attribute | The number of attribute weighting algorithms that indicated the DEGs algorithm weighting |
|---|---|
| RSU1 | 5 |
| MRPS18B | 3 |
| PPIA | 3 |
| TAGLN2 | 3 |
| ATP5B | 3 |
| VAMP8 | 3 |
| THTPA | 3 |
| FTH1 | 3 |
| RPLP2 | 3 |
| LAS1L | 3 |
| RNF126 | 3 |
| EMP3 | 3 |
| STMN1 | 3 |
| KDELR2 | 3 |
| HSPA8 | 3 |
Here, type of species (Tammar Wallaby, Rat, and Cow) was included in analysis as well as expression levels of genes. The number of attribute weighting for differentially expressed genes and organism by different attributes weighting algorithms higher than 0.5. Total number of attribute weighting algorithms which have announced the certain attribute important (weight higher than 0.5, Supplementary Data Sheet S11). This table presents the number of algorithms that selected the attribute. Weighting algorithms were Uncertainty, Gini index, Chi Squared, Rule, Information Gain, and Information Gain Ratio.