Cell transfer across ST slides of multiple time points

Multiple ST tissue sections obtained at different timepoints offer a valuable opportunity to gain a deeper understanding of cell transitions during organ development, such as mouse midbrain development and zebrafish embryogenesis. spaTrack can provide a novel strategy to infer cell tracing across multiple tissue sections based on an unbalanced and structured optimal transport algorithm by direct mapping cells among samples without integrating data. This approach is designed for cases where multiple sections containing the same cell types.In this tutorial, We use mouse midbrain ST data to demonstrate cell transfer across ST slides of multiple time points.

[1]:

import warnings

warnings.filterwarnings("ignore")

[2]:

import spaTrack as spt
import scanpy as sc
import pandas as pd
import matplotlib.pyplot as plt

sc.settings.verbosity = 0
plt.rcParams['figure.dpi'] = 200 #分辨率

1.Prepare ST data

Spatial transcriptmic data of mouse midbrain at E12.5,E14.5 and E16.5 time points were downloaded from published paper(Chen et al.2022, Cell). The example of expression tsv file and additional information file in this part can be download from Baidu Netdisk and Google Drive .We focused on progenitors cells including RGC(radial glia cells),NeuB(neuroblasts) and GlioB(glioblasts) to describe transfer relations among these progenitors cells.

[3]:

adata_all=sc.read('../../../../data/mouse.midbrain.three.time/mouse midbrain.three.time.point.ST.exp.tsv')
df_annot=pd.read_table('../../../../data/mouse.midbrain.three.time/meta.tsv',sep='\t',index_col=[0])
adata_all.obs=df_annot
adata_all.obs.index=list(df_annot.index)
adata_all.obsm['spatial']=adata_all.obs[['x','y']].values
adata_all.obs['CellID']=adata_all.obs.index

[4]:

adata_all.obs.head()

[4]:
annotation Time point x y batch CellID
CELL.100034_1 RGC E12.5 2829.0 -10000.0 0 CELL.100034_1
CELL.100795_1 RGC E12.5 2893.0 -9946.5 0 CELL.100795_1
CELL.100841_1 RGC E12.5 2672.0 -9944.0 0 CELL.100841_1
CELL.100867_1 RGC E12.5 2728.0 -9943.0 0 CELL.100867_1
CELL.100880_1 RGC E12.5 2927.0 -9939.0 0 CELL.100880_1 
[5]:

adata1_sub=adata_all[adata_all.obs['Time point']=='E12.5']
adata2_sub=adata_all[adata_all.obs['Time point']=='E14.5']
adata3_sub=adata_all[adata_all.obs['Time point']=='E16.5']
adata_all.uns['annotation_colors']=['#9A154C', '#F3754E', '#5B58A7']

Visualzation of mouse midbrain ST data of three time points.

[6]:

fig, axs = plt.subplots(ncols=1, nrows=1, figsize=(12, 6))
sc.pl.embedding(adata_all, basis='spatial', color='annotation', size=50, legend_loc='right margin', legend_fontsize=20,ax=axs, legend_fontoutline=3, show=False, s=80)
axs.set_title('spatial',fontsize=18)

[6]:

Text(0.5, 1.0, 'spatial')
../_images/notebooks_04.ST_data_of_mouse_midbrain_with_multiple_times_10_1.png

2.Calculate transition probability across sections

spaTrack infers cell tracing across multiple ST sections based on an unbalanced and structured optimal transport algorithm. We can calculate the transfer matrix between two ST sections using the spt.transfer_matrix function. Additionally, the transition probability between two sections can be calculated using spt.map_data. To obtain complete results across multiple sections, spt.generate_animate_input was used to integrate the transition probability results.

[7]:

pi12 = spt.transfer_matrix(adata1_sub,adata2_sub,epsilon=0.01,alpha=0.5)
pi23 = spt.transfer_matrix(adata2_sub,adata3_sub,epsilon=0.01,alpha=0.5)

We also can obtain the result containing cell transition probability among three ST sections.

[8]:

pi_matrix_coord = spt.generate_animate_input(pi_list=[pi12,pi23], adata_list=[adata1_sub,adata2_sub,adata3_sub],time ='Time point',annotation='annotation')
# pi_matrix_coord=pi_matrix_coord.loc[(pi_matrix_coord['pi_value1']> value1) & (pi_matrix_coord['pi_value1']> value2) ]
pi_matrix_coord.to_csv("./cell.transfer.csv")

[9]:

pi_matrix_coord.head()

[9]:
slice1 slice2 slice3 slice1_annotation slice2_annotation slice3_annotation slice1_x slice2_x slice3_x slice1_y slice2_y slice3_y k_line_12 b_line_12 k_line_23 b_line_23 pi_value1 pi_value2
0 CELL.100034_1 CELL.174282_3 CELL.6090_2 RGC GlioB RGC 2829.0 7709.0 14163.004072 -10000.0 -9508.0 -9438.505655 0.100820 -10285.218852 0.010768 -9591.007680 0.000043 0.000118
1 CELL.105423_1 CELL.174282_3 CELL.6090_2 NeuB GlioB RGC 2325.0 7709.0 14163.004072 -9678.5 -9508.0 -9438.505655 0.031668 -9752.127879 0.010768 -9591.007680 0.000076 0.000118
2 CELL.100795_1 CELL.174235_3 CELL.5906_2 RGC RGC GlioB 2893.0 8825.0 15324.653342 -9946.5 -9519.0 -9458.827390 0.072067 -10154.989127 0.009258 -9600.700247 0.000055 0.000075
3 CELL.105209_1 CELL.174235_3 CELL.5906_2 GlioB RGC GlioB 3007.0 8825.0 15324.653342 -9693.0 -9519.0 -9458.827390 0.029907 -9782.930904 0.009258 -9600.700247 0.000046 0.000075
4 CELL.114403_1 CELL.174235_3 CELL.5906_2 RGC RGC GlioB 2856.0 8825.0 15324.653342 -8887.5 -9519.0 -9458.827390 -0.105797 -8585.344865 0.009258 -9600.700247 0.000066 0.000075

The object “pi_matrix_coord” contains comprehensive results from multiple sections. In the dataframe, First three columns represent the CellID in slice1, slice2, and slice3 with the maximum transition probability among them. The pi_value1 represents the maximum transition probability value between the CellID in slice1 and slice2. Similarly, the pi_value2 represents the maximum transition probability value between the CellID in slice2 and slice3. The k_line and b_line represent slopes and intercepts of lines connecting two cells used by spt.animate_transfer() function to display dynamic figure. Users have the option to filter the maximum transition probability values based on their distribution.

3.Visualze cell transfer by dynamic figure

The cell transfer among three ST sections can be visualzed by the formation of animation

[10]:

spt.animate_transfer(adata_list=[adata1_sub,adata2_sub,adata3_sub],
                 transfer_path='./cell.transfer.csv',
                 fig_title = " Cell transfer from E12.5 E14.5 to E16.5",
                 time ='Time point',
                 color=['#5B58A7', '#9A154C', '#F3754E'],
                 annotation='annotation',
                 save_path='./cell.transfer.animation.html',
                 times=['E12.5','E14.5','E16.5'])

image0

4.Visualize cell transfer by 3D figure

Preparation of input data. We should perpare the cell id among three ST sections with line connecting.

[11]:

df_adata_1=adata1_sub.obs[['CellID','annotation','x','y']]
df_adata_1['batch']='slice1'
df_adata_2=adata2_sub.obs[['CellID','annotation','x','y']]
df_adata_2['batch']='slice2'
df_adata_3=adata3_sub.obs[['CellID','annotation','x','y']]
df_adata_3['batch']='slice3'
df_concat=pd.concat([df_adata_1,df_adata_2,df_adata_3])

df_mapping_cell=pi_matrix_coord[['slice1','slice2','slice3']]

[12]:

df_mapping_cell.head()

[12]:
slice1 slice2 slice3
0 CELL.100034_1 CELL.174282_3 CELL.6090_2
1 CELL.105423_1 CELL.174282_3 CELL.6090_2
2 CELL.100795_1 CELL.174235_3 CELL.5906_2
3 CELL.105209_1 CELL.174235_3 CELL.5906_2
4 CELL.114403_1 CELL.174235_3 CELL.5906_2

We can visualize the three sections and the line cells with the maximum transition probability in a 3D space.

[13]:

spt.visual_3D_mapping_3(df_concat,df_mapping_cell,point_size=0.8,line_width=0.06,cell_color_list=['#9A154C', '#F3754E', '#5B58A7'])
../_images/notebooks_04.ST_data_of_mouse_midbrain_with_multiple_times_27_0.png

We can visualize specific two consecutive sections after calculating the transition probability across the sections. Now,we can obtain the file containing cell transition probability between section 1(adata1_sub) and section 2(adata2_sub).

[14]:

pi_df=spt.map_data(pi12,adata1_sub,adata2_sub)

Before visualizing, users are given the option to retain higher values of the maximum transition probability values based on their distribution.

[15]:

# import seaborn as sns
# sns.displot(pi_df['pi_value'],bins=200)
# pi_df_new=pi_df.loc[pi_df['pi_value']>cutoff_value]

Visualize cell transfer between two sections by 3D

[16]:

spt.visual_3D_mapping_2(df_concat,pi_df[['slice1','slice2']],point_size=1,line_width=0.05,cell_color_list=['#9A154C', '#F3754E', '#5B58A7'])
../_images/notebooks_04.ST_data_of_mouse_midbrain_with_multiple_times_33_0.png

5.Construct gene regulation network

Here, we use adata from E12.5 and E14.5 as an example to construct regulatory network between TF genes and target genes inferred along the developmental trajectory between two time points.TF genes list of Mouse and Human can be download from the link.

[17]:

adata1_sub.write('./E12.5.adata.h5ad')
adata2_sub.write('./E14.5.adata.h5ad')

[18]:

pi_df=spt.map_data(pi12,adata1_sub,adata2_sub)
pi_df.to_csv('./cell_mapping.csv')

[19]:

gr = spt.Trainer(
    data_type='2_time',
    expression_matrix_path=['./E12.5.adata.h5ad','./E14.5.adata.h5ad'],
    cell_mapping_path='./cell_mapping.csv',
    tfs_path='../../../../data/mouse.midbrain.three.time/ms.TF.tsv',
    min_cells=[50,20],         #recommend maintaining gene expression in ～ 3% of cells.
    cell_generate_per_time=10000,
    cell_select_per_time=50
)

gr.run(training_times=10,iter_times=30)

Train 10: 100%|██████████| 300/300 [02:37<00:00,  1.90it/s, train_loss=0.0144, test_loss=0.0158]

Weight relationships of tfs and genes are stored in weights.csv.

Show the relationship between TF and target genes by scatter diagram.

[20]:

gr.plot_scatter()
../_images/notebooks_04.ST_data_of_mouse_midbrain_with_multiple_times_40_0.png

Get weights between TFs and genes.

[21]:

gr.network_df

[21]:
TF gene weight
347 Nme1 Gm20390 0.031272
950 Meis3 Ncstn 0.021513
1197 Mtf1 Nfia 0.021020
2294 Ing4 Cdc123 0.020342
1785 Mtf1 Dnajc2 0.020236
... ... ... ...
1630 Chd1 Foxk2 0.010869
435 Creb1 Cnot8 0.010869
2862 Chd1 Kdm2a 0.010868
694 Srebf1 Vps4b 0.010868
284 Apex1 Ostc 0.010867

3000 rows × 3 columns

Plot the gene regulatory network diagram according to the 'gr.network_df'.

[22]:

gr.plot_gene_regulation(min_weight=0.012,min_node_num=10)

num of weight pairs after weight filtering: 1388
num of weight pairs after node_count filtering: 394
../_images/notebooks_04.ST_data_of_mouse_midbrain_with_multiple_times_44_1.png