RNA Only Workflow¶

This is the most standard SCRIPro computation process, requiring only the input of the corresponding scRNA-seq sequencing matrix. To demonstrate SCRIP’s ability to be applied to different tissue types and infer target genes for TRs, we applied SCRIP to 10X lymphoma sequencing data. Data are available on https://www.10xgenomics.com/datasets/fresh-frozen-lymph-node-with-b-cell-lymphoma-14-k-sorted-nuclei-1-standard-2-0

Using Shell:¶

The resulting tf_score matrix can be obtained by using the following shell statement:

scripro enrich -i ./data/rna/rna.h5ad -n 50 -s hs -p rna_workflow -t 32

The resulting gata3_score matrix can be obtained through the following shell statement, where rna_workflow.pkl is the result of SCRIPro enrich:

scripro get_tf_target -i rna_workflow.pkl -t GATA3 -p GATA3_target

Using Python for custom analysis:¶

import numpy as np
import pandas as pd
import scanpy as sc
import h5py
import scripro
import anndata
import matplotlib.pyplot as plt
import numpy as np
import seaborn as sns
import pandas as pd
import scanpy as sc
import warnings
warnings.filterwarnings("ignore")

Load and preprocess data¶

rna = sc.read_h5ad('./data/rna/rna.h5ad')
rna.var_names_make_unique()
rna.raw = rna
sc.pp.normalize_total(rna, target_sum=1e4)
sc.pp.log1p(rna)
sc.pp.highly_variable_genes(rna, min_mean=0.0125, max_mean=3, min_disp=0.5)
rna = rna[:, rna.var.highly_variable]
sc.pp.scale(rna, max_value=10)
sc.tl.pca(rna, svd_solver='arpack')
sc.pp.neighbors(rna)
sc.tl.umap(rna)
sc.tl.leiden(rna)

Calculate the metacell and the marker genes corresponding to metacell.

test_data = scripro.Ori_Data(rna,Cell_num=50)

ad_all is the integrated counting matrix.

test_data.ad_all

	MIR1302-2HG	FAM138A	OR4F5	AL627309.1	AL627309.3	AL627309.2	AL627309.5	AL627309.4	AP006222.2	AL732372.1	...	AC133551.1	AC136612.1	AC136616.1	AC136616.3	AC136616.2	AC141272.1	AC023491.2	AC007325.1	AC007325.4	AC007325.2
20_0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	...	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
15_0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	...	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
15_1	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	...	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
15_2	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	...	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
13_0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	...	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
9_4	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	...	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
9_5	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	...	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
9_6	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	...	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
9_7	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	...	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
21_0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	...	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0

224 rows × 36621 columns

test_data.get_positive_marker_gene_parallel()
rna_seq_data = scripro.SCRIPro_RNA(5,'hg38',test_data,assays=['Direct','DNase','H3K27ac'])

The computational process of In Silico Deletion¶

rna_seq_data.cal_ISD_cistrome()

The P-value matrix of each metacell LISA is obtained according to the calculation results

Get TF activity Score¶

rna_seq_data.get_tf_score()
rna_seq_data.P_value_matrix

	ADNP	AFF1	AFF4	AGO1	AHR	AIRE	ALX1	ALX3	ALX4	ANHX	...	ZSCAN22	ZSCAN23	ZSCAN29	ZSCAN30	ZSCAN31	ZSCAN4	ZSCAN5A	ZSCAN5C	ZXDB	ZXDC
row
0_0	1.982159e-05	0.114342	0.466165	3.044442e-03	0.065143	0.116164	0.261117	0.090598	0.043649	0.070920	...	0.001946	1.034024e-03	0.000837	0.023628	0.187771	0.130556	0.000345	0.072917	9.929228e-07	1.078112e-06
0_1	1.078489e-03	0.045135	0.541748	4.741197e-02	0.172083	0.137448	0.120097	0.091863	0.078125	0.097334	...	0.027452	6.524492e-02	0.119130	0.071906	0.200513	0.117636	0.007210	0.072906	1.114402e-05	3.193426e-03
0_10	1.945398e-04	0.150389	0.350183	7.688059e-02	0.089623	0.316572	0.277354	0.399970	0.437044	0.195209	...	0.021498	1.736244e-03	0.091324	0.003618	0.320272	0.071882	0.000904	0.098806	2.213682e-06	1.677967e-02
0_11	9.016532e-02	0.124475	0.635978	2.211520e-02	0.178290	0.010232	0.077026	0.126848	0.065793	0.001066	...	0.211864	4.717477e-02	0.126473	0.111667	0.130438	0.169036	0.055158	0.244485	4.748398e-04	1.358551e-02
0_12	1.508612e-01	0.220131	0.714978	1.149924e-01	0.166783	0.000201	0.019816	0.003010	0.003320	0.003520	...	0.349635	1.420289e-01	0.171647	0.123673	0.080900	0.042576	0.047124	0.017884	1.611482e-01	2.017362e-01
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
9_3	1.481955e-05	0.161472	0.000004	6.475927e-07	0.004738	0.093825	0.145126	0.158836	0.204868	0.006100	...	0.000030	6.431066e-08	0.041991	0.001208	0.000560	0.012364	0.000022	0.036678	5.952748e-08	2.198499e-08
9_4	1.624109e-07	0.304159	0.185860	1.608332e-02	0.018612	0.205191	0.173053	0.138393	0.167866	0.051846	...	0.006800	1.012524e-04	0.031388	0.001566	0.097648	0.044065	0.000073	0.019923	1.451613e-03	7.308369e-03
9_5	1.541161e-06	0.252129	0.000368	4.775720e-04	0.036822	0.136602	0.147106	0.204738	0.165820	0.031218	...	0.015975	1.854799e-03	0.069004	0.008719	0.092146	0.088071	0.000901	0.005200	1.631952e-04	3.722424e-05
9_6	6.143819e-05	0.349253	0.150809	3.164199e-02	0.089277	0.122468	0.182552	0.158537	0.181882	0.090961	...	0.012562	5.747627e-03	0.085607	0.011577	0.090943	0.081455	0.004634	0.016923	3.773492e-03	5.942802e-02
9_7	6.450485e-04	0.390047	0.199128	1.675784e-02	0.132506	0.096528	0.102888	0.107414	0.135996	0.100875	...	0.016645	9.027264e-03	0.067132	0.021804	0.122074	0.053077	0.000223	0.008073	8.117502e-03	7.536773e-03

224 rows × 1226 columns

The corresponding RP score and expression value are used to weight the P-value obtained, and the final tf activity score is obtained

rna_seq_data.tf_score

	ADNP	AFF1	AFF4	AGO1	AHR	AIRE	ALX1	ALX3	ALX4	ANHX	...	ZSCAN22	ZSCAN23	ZSCAN29	ZSCAN30	ZSCAN31	ZSCAN4	ZSCAN5A	ZSCAN5C	ZXDB	ZXDC
row
0_0	1.181346e-05	0.060435	0.307493	1.462677e-04	0.026594	0.0	0.0	0.0	0.0	0.0	...	0.0	0.0	0.000038	0.001076	0.0	0.0	8.489554e-06	0.0	4.385504e-08	6.365249e-07
0_1	8.153228e-04	0.028895	0.455855	2.507957e-03	0.008484	0.0	0.0	0.0	0.0	0.0	...	0.0	0.0	0.006283	0.005021	0.0	0.0	3.823604e-03	0.0	5.890852e-07	1.917075e-03
0_10	1.138860e-04	0.095834	0.293383	3.976904e-02	0.037968	0.0	0.0	0.0	0.0	0.0	...	0.0	0.0	0.001849	0.000498	0.0	0.0	6.564053e-05	0.0	1.429377e-07	9.996831e-03
0_11	6.903511e-02	0.076661	0.422427	1.190686e-03	0.011600	0.0	0.0	0.0	0.0	0.0	...	0.0	0.0	0.006117	0.008547	0.0	0.0	4.512259e-03	0.0	3.173963e-05	8.298006e-03
0_12	8.898146e-02	0.136908	0.467959	5.825133e-02	0.009677	0.0	0.0	0.0	0.0	0.0	...	0.0	0.0	0.008715	0.010824	0.0	0.0	1.991192e-03	0.0	9.687363e-03	1.161664e-01
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
9_3	8.815053e-06	0.060870	0.000002	4.206756e-08	0.001850	0.0	0.0	0.0	0.0	0.0	...	0.0	0.0	0.000759	0.000037	0.0	0.0	6.775830e-07	0.0	2.239746e-09	1.277747e-08
9_4	1.216592e-07	0.160054	0.121961	9.350271e-04	0.007390	0.0	0.0	0.0	0.0	0.0	...	0.0	0.0	0.030882	0.000199	0.0	0.0	2.069416e-06	0.0	6.039253e-05	9.517900e-05
9_5	1.182939e-06	0.095240	0.000301	2.557181e-05	0.002016	0.0	0.0	0.0	0.0	0.0	...	0.0	0.0	0.002766	0.000738	0.0	0.0	4.742670e-04	0.0	6.393928e-06	2.261003e-05
9_6	3.662379e-05	0.186604	0.140389	2.558394e-03	0.038996	0.0	0.0	0.0	0.0	0.0	...	0.0	0.0	0.083027	0.000672	0.0	0.0	3.029751e-04	0.0	1.696993e-04	3.468848e-02
9_7	4.759191e-04	0.254694	0.129037	9.390380e-04	0.009117	0.0	0.0	0.0	0.0	0.0	...	0.0	0.0	0.063992	0.002329	0.0	0.0	1.695396e-05	0.0	8.699077e-04	4.172868e-03

224 rows × 1226 columns

Calculate the downstream target gene of each TF in each metacell¶

gata3_score = rna_seq_data.get_tf_target('GATA3')
gata3_score

	SOS1	ZNF487	PPP1CA	CFLAR	WDR37	CTLA4	STK10	NFKBIL1	INO80B	PPP2R5C	...	BCL2	RPL18	PRSS55	UBL4B	FAM13A	WDR20	SYTL3	ASH1L	APOC3	CPNE8
3_10	0.012644	0.000000	0.000000	0.000000	0.096325	0.000000	0.026573	0.000000	0.067059	0.021823	...	0	0	0	0	0	0	0	0	0	0
4_1	0.239298	0.025236	0.000000	0.111141	0.000000	0.133851	0.000000	0.000000	0.000000	0.077034	...	0	0	0	0	0	0	0	0	0	0
1_0	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	...	0	0	0	0	0	0	0	0	0	0
12_4	0.000000	0.000000	0.120566	0.209552	0.093906	0.000000	0.000000	0.000000	0.000000	0.000000	...	0	0	0	0	0	0	0	0	0	0
10_4	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	...	0	0	0	0	0	0	0	0	0	0
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
22_0	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	...	0	0	0	0	0	0	0	0	0	0
1_17	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	...	0	0	0	0	0	0	0	0	0	0
0_24	0.095861	0.000000	0.000000	0.051342	0.000000	0.000000	0.012070	0.000000	0.000000	0.000000	...	0	0	0	0	0	0	0	0	0	0
0_3	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.027181	0.220751	0.000000	0.030307	...	0	0	0	0	0	0	0	0	0	0
3_2	0.000000	0.000000	0.000000	0.000000	0.009417	0.045787	0.115108	0.000000	0.000000	0.000000	...	0	0	0	0	0	0	0	0	0	0

224 rows × 3084 columns