Differential Expression¶

AnnSQL offers simple access to the calculate_differential_expression method which performs differential expression between two groups of cells byway of a ttest. The results are stored in the diff_expression table which may be queried for results. Alternatively, the plot_differential_expression method provides a convenient way to produce a volcano plot of differential expression results.

Load AnnSQL¶

In [1]:

from AnnSQL import AnnSQL

from AnnSQL import AnnSQL

Open the AnnSQL Database¶

In [3]:

asql = AnnSQL(db="../examples/db/PCCM.asql")

asql = AnnSQL(db="../examples/db/PCCM.asql")

Take a look at the obs table¶

In [6]:

asql.query("SELECT * FROM obs LIMIT 10")

asql.query("SELECT * FROM obs LIMIT 10")

Out[6]:

	cell_id	infection_count	infected	n_genes_by_counts	total_counts	total_counts_mt	pct_counts_mt	doublet_score	predicted_doublets	n_counts	clusters	leiden_clusters	cell_type
0	TTACCGCGTGACCTGC	91	True	7586	18685.0	0.0	0.0	0.255172	False	39993.0	0	4	Glutamatergic
1	CATGCTCCACACGCCA	201	True	7314	19723.0	0.0	0.0	0.119374	False	39960.0	0	1	Glutamatergic
2	ATTGTTCAGGTCGTAG	78	True	7333	18472.0	0.0	0.0	0.081917	False	39142.0	0	4	Glutamatergic
3	TGAATCGCATCGAACT	28	True	7562	19875.0	0.0	0.0	0.322176	False	39054.0	5	9	Interneurons
4	CTCTCAGCACACCGAC	52	True	7214	18821.0	0.0	0.0	0.103203	False	38879.0	1	4	Glutamatergic
5	AAGCCATCATTAAGCC	321	True	7240	18956.0	0.0	0.0	0.146727	False	38869.0	1	1	Glutamatergic
6	CTCAATTTCTCTCGAC	72	True	7256	18034.0	0.0	0.0	0.062750	False	38789.0	0	2	Glutamatergic
7	TTCATGTCATCCTGTC	148	True	7118	18250.0	0.0	0.0	0.163814	False	38704.0	1	4	Glutamatergic
8	AGATCCATCCATCGTC	46	True	7738	18869.0	0.0	0.0	0.351351	False	38675.0	0	4	Glutamatergic
9	CAATTTCCACTTGAAC	202	True	7197	18149.0	0.0	0.0	0.108257	False	38632.0	0	0	Glutamatergic

Calculate Differential Expression¶

Here we calculate differential expression between two leiden clusters determined previously.

In [7]:

asql.calculate_differential_expression(obs_key="leiden_clusters", group1_value="0", group2_value="9", drop_table=False, name="leiden_0_vs_9")

asql.calculate_differential_expression(obs_key="leiden_clusters", group1_value="0", group2_value="9", drop_table=False, name="leiden_0_vs_9")

DE Calculation Complete.

Plot the differential expression¶

In [8]:

asql.plot_differential_expression(pvalue_threshold=0.01, logfc_threshold=1, group1="0", group2="9")

asql.plot_differential_expression(pvalue_threshold=0.01, logfc_threshold=1, group1="0", group2="9")

Query the results with:
"SELECT * FROM diff_expression WHERE group1='0' and group2='9'".

Query the results from the diff_expression table¶

In [13]:

asql.query("SELECT * FROM diff_expression WHERE group1='0' and group2='9' and name='leiden_0_vs_9'")

asql.query("SELECT * FROM diff_expression WHERE group1='0' and group2='9' and name='leiden_0_vs_9'")

Out[13]:

	name	group1	group2	gene	tstat	logfc	df	pval	adj_pval
0	leiden_0_vs_9	0	9	n0610007P14Rik	9.085450	0.626459	233.578499	0.000000e+00	8.756260e-10
1	leiden_0_vs_9	0	9	n0610010F05Rik	1.918118	0.109200	238.122874	5.629211e-02	7.830868e-02
2	leiden_0_vs_9	0	9	n1110001J03Rik	-6.253281	-0.472151	231.805321	1.915299e-09	1.845523e-10
3	leiden_0_vs_9	0	9	n1110008F13Rik	-0.268244	-0.016895	247.469853	7.887353e-01	8.197332e-08
4	leiden_0_vs_9	0	9	n1110008P14Rik	3.425778	0.258434	244.577604	7.188917e-04	3.164319e-08
...	...	...	...	...	...	...	...	...	...
2495	leiden_0_vs_9	0	9	Zrsr2	-3.958356	-0.272130	230.565107	1.005430e-04	7.165040e-01
2496	leiden_0_vs_9	0	9	Zyg11b	2.307880	0.126168	221.508122	2.192866e-02	0.000000e+00
2497	leiden_0_vs_9	0	9	Zzef1	-1.241593	-0.111361	247.840197	2.155605e-01	6.358348e-06
2498	leiden_0_vs_9	0	9	l7Rn6	0.382926	0.033347	220.552533	7.021432e-01	1.506674e-01
2499	leiden_0_vs_9	0	9	mt_Nd6	-1.390571	-0.146923	230.419130	1.656971e-01	3.225584e-01

2500 rows × 9 columns