---
title: "DGE along pseudotime"
author: "Luise A. Seeker"
date: "19/04/2021"
output: html_document
---
#### DGE along pseudotime
```{r, figure.align = "center", fig.width = 6, fig.height = 6}
plot_heatmap(
model_slingshot,
expression_source = my_dat_ba4$expression,
grouping = ba4$ol_clusters_named,
features_oi = 50
)
```
##### Lineage/branch markers
We can also extract features specific for a branch, eg. genes which change when
a cell differentiates into a specific oligo subtype
##### Oligo A
```{r, figure.align = "center", fig.width = 6, fig.height = 6}
branch_feature_importance <-
calculate_branch_feature_importance(model_slingshot,
expression_source=my_dat_ba4$expression)
oligo_A_features <- branch_feature_importance %>%
filter(to == which(model_slingshot$milestone_labelling =="Oligo_A")) %>%
top_n(50, importance) %>%
pull(feature_id)
plot_heatmap(
model_slingshot,
expression_source = my_dat_ba4$expression,
features_oi = oligo_A_features
)
```
##### Oligo C
```{r, figure.align = "center", fig.width = 6, fig.height = 6}
oligo_F_features <- branch_feature_importance %>%
filter(to == which(model_slingshot$milestone_labelling =="Oligo_F")) %>%
top_n(50, importance) %>%
pull(feature_id)
plot_heatmap(
model_slingshot,
expression_source = my_dat_ba4$expression,
features_oi = oligo_F_features
)
```
# SCORPIUS
```{r}
set.seed(1)
model <- infer_trajectory(my_dat_ba4, "scorpius")
```
```{r}
plot_dimred(model, dimred = my_dat_ba4$dimred)
```
# PAGA
```{r}
set.seed(1)
model = dynwrap::infer_trajectory(my_dat_ba4, ti_paga(filter_features = F),
verbose = TRUE)
```
```{r}
plot_dimred(model)
```
```{r}
plot_dimred(model, dimred = my_dat_ba4$dimred)
```
# PAGA TREE
```{r}
set.seed(1)
model_paga_tree = dynwrap::infer_trajectory(my_dat_ba4, ti_paga_tree(filter_features = F),
verbose = TRUE)
```
```{r}
plot_dimred(model_paga_tree)
```
```{r}
plot_dimred(model_paga_tree, dimred = my_dat_ba4$dimred)
```
# MST
```{r}
set.seed(1)
model_mst = dynwrap::infer_trajectory(my_dat_ba4, ti_mst())
```
```{r}
plot_dimred(model_mst)
```
```{r}
plot_dimred(model_mst, dimred = my_dat_ba4$dimred)
```
```{r, fig.width = 6, fig.height=6, fullwidth = TRUE}
patchwork::wrap_plots(
plot_dimred(model_mst, dimred = my_dat_ba4$dimred) + ggtitle("Cell ordering"),
plot_dimred(model_mst, dimred = my_dat_ba4$dimred, grouping = group_onto_nearest_milestones(model_mst)) + ggtitle("Cell grouping"),
plot_dimred(model_mst, dimred = my_dat_ba4$dimred, feature_oi = "SPARC", expression_source = my_dat_ba4) + ggtitle("Feature expression"),
plot_dimred(model_mst, dimred = my_dat_ba4$dimred, "pseudotime", pseudotime = calculate_pseudotime(model_mst)) + ggtitle("Pseudotime")
)
```
```{r}
plot_dimred(model_mst, dimred = my_dat_ba4$dimred, grouping = group_onto_nearest_milestones(model_mst)) + ggtitle("Cell grouping")
```
```{r}
model_rooted_mst <- model_mst %>% add_root(root_milestone_id = "M14")
```
```{r}
plot_dimred(model_rooted_mst, dimred = my_dat_ba4$dimred,
"pseudotime",
pseudotime = calculate_pseudotime(model_mst)) +
ggtitle("Pseudotime")
```
```{r}
plot_dendro(model_mst)
```
```{r}
plot_heatmap(model_mst, expression_source = my_dat_ba4)
```
```{r}
overall_feature_importances <-
dynfeature::calculate_overall_feature_importance(model_mst, expression_source = my_dat_ba4)
features <- overall_feature_importances %>%
top_n(40, importance) %>%
pull(feature_id)
```
```{r}
plot_graph(my_dat_ba4)
```
# Save files
```{r}
ti_save_dir <- here("outs", "traj_inf_output")
dir.create(ti_save_dir)
saveRDS(model_paga_tree, here("outs", "traj_inf_output", "model_paga_tree_ba4.RDS"))
```