Citizen science helps to raise awareness about gut microbiome health in people at risk of developing non-communicable diseases

Figures & data

Table 1. Participants’ demographics and nutritional parameters obtained after analysis of the food registry and GPAQ questionnaires.

Sociodemographic characteristics		N = 70^a
Age (years)		38 (21–58)
Educational levelb	Secondary education	7(10%)
	Tertiary education	3(4.3%)
	Bachelor’s or equivalent	20(29%)
	Master o superior	40(57%)
Residence	Madrid region	47(67%)
	Other communities	23(33%)
Sex	Male	12(17%)
	Female	58(83%)
BMI (kg/m2	Normal range	47(67%)
	Overweight/Obese	23(33%)
Smoking		14(20%)
Physical activityc	Low	17(24%)
	Medium	23(33%)
	High	30(43%)
Healthy Eating Indexd	Poor	12(19.4%)
	Fair	15(24.1%)
	Good	18(29%)
	Very Good	12(19.4)
	Excellent	5(8.1%)
	NA	8

^amean (range); n(%). ^b According to the International Standard Classification of Education. ^c according to the GPAQ questionnaire.^35,d According to Healthy Eating Index.³⁷

Figure 1. (a) Patient similarity network constructed from Aitchison distance and volunteers’ clusters, (b) Proportion of pictures provided to each photovoice category by participants belonging to each cluster.

1a) Patient similarity network constructed with the Aitchison distance. Each node represents a patient, and the colors of the nodes indicate the clusters they belong to after applying k-nn clustering: cluster one (purple), cluster two (pink), and cluster three (yellow). Edges indicate the similarity between patients calculated using the Aitchison distance. The thicker the edge, the more similar the patients are. 1b) Proportions of photographs cataloged to each photovoice theme (Balance, Foodie, Mindful eating, Wellness) according to clusters.

Table 2. Participants’ demographics, lifestyle parameters and microbiome-knowledge results obtained after analysis of the food registry, GPAQ and ad-hoc microbiome-knowledge questionnaires according to the three clusters identified with patient similarity network.

Characteristic^a		Cluster 1 (n = 36)	Cluster 2 (n = 27)	Cluster 3 (n = 7)	P value^g
Age (years)		37	39	34	0.8
Gender	Men	8 (22%)	3 (11%)	1 (14%)	0.5
	Women	28 (78%)	24 (89%)	6 (86%)
BMI (kg/m2)		25.3	23.3	21.2	0.060
Weight (kg)		69	63	56	0.038
F/B^b ratio		1.47	1.45	1.10	0.8
Physical activity^c	Low	10 (28%)	6 (22%)	1 (14%)	0.9
	Medium	10 (28%)	10 (37%)	3 (43%)
	High	16 (44%)	11 (41%)	3 (43%)
Healthy eating index^d	Poor	8(26%)	3(12%)	1(14%)	0.3
	Fair	7(23%)	6(25%)	2(29%)
	Good	6(19%)	10(42%)	2(29%)
	Very Good	5(16%)	5(21%)	2(29%)
	Excellent	5(16%)	0(0%)	0(0%)
Smoking		5(14%)	7(26%)	2(29%)	0.4
Two or more risk factors for NCD’s		13(36%)	7(26%)	2(29%)	0.8
Initial knowledge score^e		13.1	13.2	14.6	0.012
Final knowledge score^e		14	14	14.3	0.7
Variation in scoring^f		0.89	0.83	−0.28	0.03

^amean; n(%). ^b Firmicutes/Bacteroidetes ratio. ^cPhysical activity according to the GPAQ questionnaire.^{35 d}According to Healthy Eating Index.^{37 e}Reported by the ad-hoc knowledge questionnaire. ^fVariation in punctuation between final and initial knowledge questionnaire. ^gKruskall-Wallis rank sum test; Fisher’s exact test. In bold are shown those variables with a significant p-value.

Figure 2. (a) Co-occurrence networks of the microbiome of each cluster of patients. (b) Alfa diversity indexes per cluster (c) Relative abundance of the top 20 genera according to each cluster.

2a) Co-occurrence network of each cluster. The nodes in the graph represent different bacterial genera. The size of each node corresponds to its relative abundance, while the edges represent statistically significant associations between the nodes (P < 0.05). Green edges indicate positive relationships, while red edges indicate negative ones. The thickness of each edge indicates the strength of the association. Nodes highlighted in figure a correspond to keystone taxa calculated as nodes with betweenness and degree greater than Quantile 0.90. 2b) Alpha diversity indexes (Chao1, Shannon, and Simpson) from each cluster *** P < 0.001, ** P < 0.01, * P < 0.05.

Table 3. Network topological properties of microbial co-occurrence networks from each cluster.

Network properties	Cluster 1	Cluster 2	Cluster 3
Number of nodes	145	184	207
Number of edges	217	275	425
Average path length	4.011	4.55	3.21
Edge density	0.018	0.016	0.019
Average degree	1.49	1.49	2.05
Positive edge percentage	91.70	96	81.88
Keystone taxaa	[Eubacterium] brachy group Candidatus Soleaferrea Lachnospiraceae UCG-006 Marvinbryantia	Pseudoflavonifractor Victivallis	Lachnospiraceae ND3007 group [Eubacterium] coprostanoligenes group Defluviitaleaceae UCG-011 Brevundimonas Coprobacter UCG-004

^aKeystone taxa calculated as nodes with betweenness and degree greater than Quantile 0.90.

Figure 3. Results of the satisfaction and learning questionnaire presented on a Likert scale from 1 (strongly disagree) to 5 (strongly agree).

Kurtz ZD, Müller CL, Miraldi ER, Littman DR, Blaser MJ, Bonneau RA, von Mering C. Sparse and compositionally robust inference of microbial ecological networks. PLoS Comput Biol. 2015;11(5):e1004226. doi:10.1371/journal.pcbi.1004226.

 PubMed Web of Science ®Google Scholar

Caballero SG, Ordoñez S, Fuentes G, Zambrano LJM. Photobook #PictureYourMicrobes. [Internet]. 2021. doi:10.5281/zenodo.5682942.

 Google Scholar