| Autor: |
Luli Zou, Erdos, Michael, D. Leland Taylor, Chines, Peter, Varshney, Arushi, Parker, Stephen, Collins, Francis, Didion, John |
| Rok vydání: |
2018 |
| DOI: |
10.6084/m9.figshare.6338675.v1 |
| Popis: |
Figure S1. Distribution of WGBS missingness across chromatin states, not normalized for total number of CpGs in that chromatin state. Figure S2. Distribution of distance from a WGBS or EPIC CpG to the nearest WGBS or EPIC CpG. Figure S3. CpG methylation pairwise differences as a function of distance genome-wide and in regions of higher across-tissue variance. Figure S4. CpG methylation pairwise differences for pancreatic islets as a function of distance genome-wide and in regions of higher across-tissue variance. Figure S5. Comparison of all CpGs and CpGs that had higher across-tissue variance. Figure S6. Joyplot showing distribution of WGBS methylation (beta) values within each chromatin state for all tissues. Figure S7. Performance at intermediate CpGs does not improve when using a balanced training distribution. Figure S8. Smooth scatterplot of beta values of CpGs shared between EPIC and WGBS. Figure S9. Imputation mitigates discordance between WGBS at EPIC at low WGBS depth regardless of EPIC probe type. Figure S10. Correlation among the top 30 features ranked by BoostMe. Figure S11. Correlation among the top 30 features ranked by random forests. Figure S12. Distribution of across-sample CpG variance values vs. number of missing values for each CpG. Table S1. Summary of the data used in this work. Table S2. Previously reported imputation metrics and those reported in this work. Table S3. All features included in BoostMe and random forests and their source. Table S4. Genome-wide performance of algorithms, trained on 500,000 CpGs, for predicting methylation values. Table S5. RMSE performance of BoostMe and random forests improves when training on continuous values. Table S6. Top 100 transcription factors ranked in descending order as reported by BoostMe, trained only using TFBS features. Table S7. Top 100 transcription factors ranked in descending order as reported by random forests trained only using TFBS features. (PDF 5703Â kb) |
| Databáze: |
OpenAIRE |
| Externí odkaz: |
|
