The Integrated Bioinformatic Assay of Genetic Expression Features and Analyses of Traditional Chinese Medicine Specific Constitution Reveal Metabolic Characteristics and Targets in Steatosis of Nonalcoholic Fatty Liver Disease

Figures & data

Table 1 The Correlational Analysis Shows Implications of Age, BMI, Blood Lipids, and Area of Liver Steatosis on Fasting Insulin or HOMA-IR

	Age	BMI	TG	TC	LDL	HDL	Steatosis
Fasting insulin	0.201	0.733**	0.554*	0.355*	0.361*	−0.353*	0.715**
HOMA-IR	0.284	0.691**	0.540*	0.359*	0.359*	−0.320	0.724**

Note: * p < 0.05, **p < 0.01.

Abbreviations: BMI, body mass index; TG, total triglyceride; TC, total cholesterol; LDL, low-density lipoprotein; HDL, high-density lipoprotein; HOMA-IR, homeostasis model assessment-insulin resistant.

Figure 1 The comparison of clinical characteristics of patients with simple steatosis (SS) versus normal health control (HC) in GSE89632; the result of blood lipid assessments showed no significance between SS and HC.

Notes: *The group of SS was compared with the HC group, p < 0.05, **p < 0.01; The data in figure steatosis use unpaired t-test; The data in figure HOMA-IR use Mann–Whitney test; The data in figure Lipid and Liver UFAs use two-way ANOVA, Sidak multiple-comparison test.

Abbreviations: TG, total triglyceride; TC, total cholesterol; LDL, low-density lipoprotein; HDL, high-density lipoprotein; UFAs, unsaturated fatty acids; ARA, arachidonic acid; EPA, eicosapentaenoic acid; DHA, docosahexaenoic acid; HOMA-IR, Homeostasis model assessment-insulin resistant; ns, not significant.

Figure 2 The Pearson correlation of clinical characteristics of NAFLD and regular liver donors. (A) Moreover, (B) shows multiple linear regression as fasting insulin is a dependent variable. (C) Furthermore, (D) shows Pearson correlation analysis of fasting insulin and HOMA-IR with multiple clinical characteristics.

Abbreviations: BMI, body mass index; TG, total triglyceride; TC, total cholesterol; LDL, low-density lipoprotein; HDL, high-density lipoprotein; HOMA-IR, homeostasis model assessment-insulin resistant.

Figure 3 Continued.

Figure 3 The bioinformatics analyses of global genetic features of NAFLD using data of GSE89632. (A) The heatmap shows the expressions of the top 50 significant genes in liver samples of SS versus HC; The deep red color represents the extent of high expressions of DEGs, whereas the green color represents the lower expressions of DEGs. The sample number from GSM238570 to GSM2385755 on the left bottom of heatmap and GSM 2385773, GSM 2385777 indicate SS group, whereas the sample numbers on the right bottom from GSM2385757 to GSM2385782 except GSM 2385773, GSM 2385777 indicate HC group. The right side lists the official gene symbols of the top 50 significant genes. (B) The volcano plot of expressions of DEGs; the define and cut-off of up- (red) or down- (green) regulated genes are the absolute values of log₂Fold change which exceed 2, and the label of Y-axis is the -log₁₀ adjusted P-value. (C) The overview of GO analysis of the DEGs, the order was defined by collected counts of DEGs in multiple Gene ontology (biological process, molecular function, and cellular component). P < 0.05. (C) The overview of KEGG pathway analysis of the DEGs, the order was defined by collected counts of DEGs in multiple KEGG pathways. P < 0.05. (D) The overview of respective GO component analysis of the DEGs, the order was defined by collected counts of DEGs in multiple Gene ontology (biological process, molecular function, and cellular component). P < 0.05. (E) WGCNA analyses show the dendrogram of genes divided into 7 modules with different colors. The height shows the heterogeneities of different co-expressed genes groups. (F) WGCNA analyses show the heatmap on the expression of 7 different modules, the different modules in different colors indicate corresponding co-expressed gene groups, the gene amounts have been indicated on the correct upper box. The bottom displays a liver sample. SS: simple Steatosis, HC: normal health control. (G) WGCNA analyses show the network connections in each meaningful module. A different color indicated each module. (H) GSEA analyzed the heatmap of top 50 genes; The upper label showed numbers of liver samples, and the correct label indicates the official gene symbols of the top 50 significantly expressed genes. Red color indicates up-regulated, and blue indicates down-regulated. The samples ID with SS indicate SS group, whereas sample ID with HC indicates HC group. (I) An overview of up-regulated genes by GSEA analysis. In each figure, the up-regulated genes sets were displayed with a positive enrichment score. The last figure of the butterfly plot showed a signal-to-noise score in this GSEA analysis.

Figure 4 Continued.

Figure 4 The bioinformatics analyses of the features of LncRNA in NAFLD using data of GSE107231. (A) The heatmap shows the expressions of top 50 significant genes including DEGs in liver samples of NAFLD versus normal; The deep red color represents the extent of high expressions of DEGs, whereas the green color represents the lower expressions of DEGs. The sample numbers from GSM2863101 to GSM2863105 on the left bottom of the heatmap indicate the normal control group, whereas the sample numbers on the right bottom from GSM2863106 to GSM2863110 indicate the NAFLD group; The right side lists the official gene symbols of significant genes. (B) The volcano plot of expressions of DEGs; the define and cut-off of up- (red) or down- (green) regulated genes are the absolute values of log₂Fold change which exceed 1, and the label of the Y-axis is the -log₁₀ adjusted P-value. (C) The overview of GO analysis of the DEGs, the order was defined by collected counts of DEGs in multiple Gene ontology (biological process, molecular function, and cellular component). P < 0.05. (D) The overview of KEGG pathway analysis of the DEGs, the order was defined by collected counts of DEGs in multiple KEGG pathways. P < 0.05. (E) The dendrogram of genes are divided into 7 modules with different colors. The height shows the heterogeneities of different co-expressed genes groups. (F) The heatmap on the expression of 10 different modules, the different modules in different colors indicate corresponding co-expressed gene groups, the gene amounts have been indicated on the upper right box. The sample numbers from GSM2863101 to GSM2863105 on the left bottom of the heatmap indicate a normal group, whereas sample numbers on the right bottom from GSM2863106 to GSM2863110 indicate the NAFLD group. (G) The network connections in each meaningful module. A different color indicated each module. (H) GSEA analyzed the heatmap of top 50 genes; The upper label showed numbers of liver samples, and the right label indicates the official gene symbols of the top 50 significantly expressed genes. Red color indicates up-regulated, and blue indicates down-regulated. The sample numbers from GSM2863101 to GSM2863105 marketed by yellow indicate a normal control group, whereas sample numbers from GSM2863106 to GSM2863110 indicate the NAFLD group. (I) An overview of up-regulated genes by GSEA analysis. In each figure, the up-regulated genes sets were displayed with a positive enrichment score. The last figure of the butterfly plot showed a signal-to-noise score in this GSEA analysis.

Figure 5 The bioinformatics analyses of the features of MiRNA in NAFLD using data of GSE114923. (A) The heatmap shows the expressions of top 455 DEGs (partially displayed) in serum samples of NAFLD versus normal; The deep red color represents the extent of high expressions of DEGs, whereas the green color represents the lower expressions of DEGs. The sample numbers from GSM3154851 to GSM3154854 on the left bottom of the heatmap indicate the NAFLD group, whereas the sample numbers on the right bottom from GSM3154855 to GSM3154850 indicate normal control group; The right side lists the official gene symbols of top 455 DEGs (partially displayed). (B) The MA plot (left) and a volcano plot (right) of expressions of DEGs; the define and cut-off of up- (red) or down- (blue) regulated genes are the absolute values of log₂Fold change which exceed 2.

Table 2 Analysis of General Characteristics and Biochemical Test Data of NAFLD and Control Group

	Age (year)	BMI (kg/m^2)	WC (cm)	WHR	SBP (mmHg)	DBP (mmHg)	FPG (mmol/l)	ALT (U/L)	AST (U/L)	TG (mmol/l)	TC (mmol/l)
Control	68.7±9.8	23.6±3.1	83.6±13.9	0.89±0.2	131.4±15.4	78.0±9.0	6.0±1.4	26.8±12.3	22.6±8.5	1.3±1.2	5.1±1.0
NAFLD	67.6±9.2	27.6±3.3**	93.3±8.0**	0.94±0.1**	136.4±14.8	80.6±8.8**	6.8±2.0**	35.2±18.0**	24.9±12.0**	2.0±1.9**	5.3±1.1**

Note: **NAFLD group compared with the control group, P < 0.01.

Abbreviations: NAFLD, non-alcoholic fatty liver disease group; BMI, body mass index; WC, waist circumference; WHR, waist-to-hip ratio; SBP, systolic blood pressure; DBP, diastolic blood pressure; FPG, fasting plasma glucose; GPT, ALT, glutamic pyruvic transaminase; AST, glutamic oxaloacetic transaminase; TG, triglyceride; TC, total cholesterol.

Figure 6 Comparisons of general characteristics and biochemical test data between NAFLD and control group on Age (A), BMI (B), waist (C), WHR (D), SBP, DBP (E), FPG (F), blood lipids (G), and aminotransferase (H); The results showed that besides Age, SBP, there were significant disparities in BMI, waist, WHR, DBP, FPG, blood lipids, and aminotransferase between the NAFLD and normal participants.

Notes: *NAFLD group compared with the control group, P < 0.05; **NAFLD group compared with the control group, P < 0.01.

Abbreviations: ns, no significance; NAFLD, non-alcoholic fatty liver disease group; BMI, Body Mass Index; WHR, Waist-to-hip ratio; FPG, fasting plasma glucose; SBP, Systolic blood pressure; DBP, Diastolic blood pressure; TG, triglyceride; TC, total cholesterol; ALT, glutamic pyruvic transaminase; AST, glutamic oxaloacetic transaminase.

Table 3 Chi-Square Analysis of Constitutional Characteristics of NAFLD and Control Group

	YYH	QAC	QSC	DHC	PDC	BSC	YAAC	YAC	ESP
Control	1934	229	81	38	19	18	499	181	28
NAFLD	1836**	226	111*	53	73***	17	333***	199	32
Chi-square	9.09	0.022	4.866	2.516	32.26	0.029	39.37	0.919	0.27
P	0.003	0.883	0.027	0.113	0.001	0.865	0.001	0.338	0.603

Notes: *NAFLD group compare with control group, P < 0.05; **P < 0.01; ***P < 0.001.

Abbreviations: YYH, yin-yang harmony; QAC, qi asthenia constitution; QSC, qi-stagnation constitution; DHC, damp-heat constitution; PDC, phlegm and dampness constitution; BSC, blood stasis constitution; YAAC, yang asthenia constitution; YAC, yin asthenia constitution; ESP, especially sensitive physique.

Figure 7 The comparisons of TCM constitution distributions between NAFLD and control groups.

Notes: *NAFLD group compare with control group, P < 0.05; **P < 0.01; ***P < 0.001.

Abbreviations: YYH, yin-yang harmony; QAC, qi asthenia constitution; QSC, qi-stagnation constitution; DHC, damp-heat constitution; PDC, phlegm and dampness constitution; BSC, blood stasis constitution; YAAC, yang asthenia constitution; YAC, yin asthenia constitution; ESP, especially sensitive physique.

Table 4 Binary Logistic Regression Analyzes the Implications of Metabolic Characteristics and TCM Constitutions on NAFLD

Variable	Age	Gender	BMI	Waist	DM	HBP	FPG	ALT	AST	TC
B	−0.005	−0.755*	0.266*	0.06*	−1.024*	−0.166*	0.138*	0.042*	−0.022*	0.058
SE	0.005	0.079	0.019	0.007	0.123	0.077	0.026	0.005	0.006	0.038
Variable	TG	YYH	QAC	YAAC	YAC	PDC	DHC	BSC	QSC	ESP
B	0.311*	0.909*	0.495*	0.899*	0.565*	0.352	0.28	0.702	0.038	−0.143
SE	0.036	0.144	0.156	0.146	0.163	0.356	0.296	0.492	0.210	0.348

Note: *p < 0.05.

Abbreviations: SE, standard error; DM, diabetes mellitus; HBP, hypertensive diseases; YYH, yin-yang harmony; QAC, qi asthenia constitution; YAAC, yang asthenia constitution; YAC, yin asthenia constitution; PDC, phlegm and dampness constitution; DHC, damp-heat constitution; BSC, blood stasis constitution; QSC, qi-stagnation constitution; ESP, Especially sensitive physique; ALT, glutamic pyruvic transaminase; AST, glutamic oxaloacetic transaminase; TG, triglyceride; TC, total cholesterol.

Table 5 Multiple Linear Regression Analysis Shows Significant Factors with Fasting Plasma Glucose (FPG) (P < 0.05)

Variable	NAFLD	Diabetes	BMI	Waist	TC	TG
B	0.213	2.013	−0.025	0.017	0.107	0.086
SE	0.052	0.060	0.010	0.004	0.021	0.013
Variable	SBP	DBP	ALT	AST	YAC
B	0.013	−0.012	0.014	−0.018	0.357
SE	0.002	0.003	0.002	0.003	0.081

Abbreviations: NAFLD, non-alcoholic fatty liver disease; ALT, glutamic pyruvic transaminase; AST, glutamic oxaloacetic transaminase; SBP, systolic blood pressure; DBP, diastolic blood pressure; TG, triglyceride; TC, total cholesterol; YAC, yin asthenia constitution.

Table 6 Binary Logistic Regression Analysis on the Implications of Metabolic Characteristics and TCM Constitutions on DM

Variable	Age	Gender	BMI	Waist	NAFLD	HBP	FPG	ALT	AST	TC
B	0.044*	−0.037	0.029	−0.017*	−0.966*	−0.349*	0.599*	<0.000	−0.002	−0.077
SE	0.006	0.098	0.020	0.008	0.117	0.098	0.028	0.005	0.006	0.043
Variable	TG	YYH	QAC	YAAC	YAC	PDC	DHC	BSC	QSC	ESP
B	−0.011	−0.058	−0.306	−0.071	−0.342*	−0.362	0.070	−0.273	−0.105	−0.433
SE	0.025	0.161	0.163	0.161	0.172	0.314	0.334	0.467	0.230	0.406

Note: *p < 0.05.

Abbreviations: SE, standard error; DM, diabetes mellitus; NAFLD, non-alcoholic fatty liver disease; HBP, hypertensive diseases; YYH, yin-yang harmony; QAC, qi asthenia constitution; YAAC, yang asthenia constitution; YAC, yin asthenia constitution; PDC, phlegm and dampness constitution; DHC, damp-heat constitution; BSC, blood stasis constitution; QSC, qi-stagnation constitution; ESP, Especially sensitive physique; ALT, glutamic pyruvic transaminase; AST, glutamic oxaloacetic transaminase; TG, triglyceride; TC, total cholesterol.

Data Sharing Statement

The data and material are conserved by Dr. Chengying Gu and can be accessed via contact the Email of Dr. Chengying Gu.