Analysis of the tumors development after grafting

Transmissible tumors dataset

Random effects selection

m1 <- glmmTMB(data = donor_trans, Tumors ~ donor * donor_status + receiver + (1 |
    lot) + (1 | date_draft), family = binomial, REML = T)
m2 <- glmmTMB(data = donor_trans, Tumors ~ donor * donor_status + receiver + (1 |
    date_draft/lot), family = binomial, REML = T)
m3 <- glmmTMB(data = donor_trans, Tumors ~ donor * donor_status + receiver + (1 |
    lot), family = binomial, REML = T)
m4 <- glmmTMB(data = donor_trans, Tumors ~ donor * donor_status + receiver + (1 |
    date_draft), family = binomial, REML = T)
m5 <- glmmTMB(data = donor_trans, Tumors ~ donor * donor_status + receiver, family = binomial,
    REML = T)

AICc(m1, m2, m3, m4, m5)

##    df     AICc
## m1  7 201.3000
## m2  7 201.3000
## m3  6 199.1171
## m4  6 199.2866
## m5  5 197.5273

There is no need to include any of the potential random effects that have been measured.

Fixed effects selection

m1 <- glmmTMB(data = donor_trans, Tumors ~ donor * donor_status * receiver, family = binomial,
    REML = F)
m2 <- glmmTMB(data = donor_trans, Tumors ~ donor * donor_status + receiver, family = binomial,
    REML = F)
m3 <- glmmTMB(data = donor_trans, Tumors ~ donor + donor_status * receiver, family = binomial,
    REML = F)
m4 <- glmmTMB(data = donor_trans, Tumors ~ donor * receiver + donor_status, family = binomial,
    REML = F)

m5 <- glmmTMB(data = donor_trans, Tumors ~ donor + donor_status + receiver, family = binomial,
    REML = F)
m6 <- glmmTMB(data = donor_trans, Tumors ~ donor + donor_status, family = binomial,
    REML = F)
m7 <- glmmTMB(data = donor_trans, Tumors ~ donor + receiver, family = binomial, REML = F)
m8 <- glmmTMB(data = donor_trans, Tumors ~ donor_status + receiver, family = binomial,
    REML = F)

m9 <- glmmTMB(data = donor_trans, Tumors ~ donor * donor_status, family = binomial,
    REML = F)
m10 <- glmmTMB(data = donor_trans, Tumors ~ donor * receiver, family = binomial,
    REML = F)
m11 <- glmmTMB(data = donor_trans, Tumors ~ donor_status * receiver, family = binomial,
    REML = F)

m12 <- glmmTMB(data = donor_trans, Tumors ~ donor, family = binomial, REML = F)
m13 <- glmmTMB(data = donor_trans, Tumors ~ donor_status, family = binomial, REML = F)
m14 <- glmmTMB(data = donor_trans, Tumors ~ receiver, family = binomial, REML = F)

m15 <- glmmTMB(data = donor_trans, Tumors ~ 1, family = binomial, REML = F)

AICc(m1, m2, m3, m4, m5, m6, m7, m8, m9, m10, m11, m12, m13, m14, m15)

##     df     AICc
## m1   8 198.1611
## m2   5 196.7301
## m3   5 200.3366
## m4   5 200.1467
## m5   4 198.2827
## m6   3 196.2356
## m7   3 214.0467
## m8   3 215.8551
## m9   4 194.6361
## m10  4 216.0085
## m11  4 217.9357
## m12  2 212.3960
## m13  2 213.8079
## m14  2 228.9173
## m15  1 226.9319

Table of the results of the best fitted models (lower AICc+2)

	Tumors			Tumors
Predictors	Odds Ratios	CI	p	Odds Ratios	CI	p
donor [Rob]	0.38	0.15 – 0.96	0.041	0.21	0.10 – 0.44	<0.001
donor status [T]	9.62	3.00 – 30.83	<0.001	4.47	2.17 – 9.22	<0.001
donor [Rob] × donor status [T]	0.23	0.05 – 1.07	0.061
Observations	164			164

simulateResiduals(m9, plot = T)

Final model results:

m9 <- glmmTMB(data = donor_trans, Tumors ~ donor * donor_status, family = binomial,
    REML = T)
tab_model(m9, show.intercept = F, show.r2 = F, show.re.var = F)

	Tumors
Predictors	Odds Ratios	CI	p
donor [Rob]	0.38	0.15 – 0.96	0.041
donor status [T]	9.62	3.00 – 30.83	<0.001
donor [Rob] × donor status [T]	0.23	0.05 – 1.07	0.061
Observations	164

There is an slightly significant effect of the status of the donor, with individuals grafted with Robusta tissues developing less tumors than healthy one. And a strong effect of the donor status, with 10,4 % (IRR of 9,62) more chances to develop tumors when the giver hydra is tumorous. Maybe a small interaction but not enough power to be sure.

Spontaneaous tumors dataset

Random effects selection

m1 <- glmmTMB(data = donor_spont, Tumors ~ donor + receiver + donor_status + (1 |
    lot) + (1 | date_draft), family = binomial, REML = T)
m2 <- glmmTMB(data = donor_spont, Tumors ~ donor + donor_status + receiver + (1 |
    date_draft/lot), family = binomial, REML = T)
m3 <- glmmTMB(data = donor_spont, Tumors ~ donor + donor_status + receiver + (1 |
    lot), family = binomial, REML = T)
m4 <- glmmTMB(data = donor_spont, Tumors ~ donor + donor_status + receiver + (1 |
    date_draft), family = binomial, REML = T)
m5 <- glmmTMB(data = donor_spont, Tumors ~ donor + donor_status + receiver, family = binomial,
    REML = T)

AICc(m1, m2, m3, m4, m5)

##    df     AICc
## m1  6 147.4359
## m2  6 147.4359
## m3  5 145.1822
## m4  5 145.2685
## m5  4 143.2151

There is no need to include any of the potential random effects that have been measured.

Fixed effects selection

m1 <- glmmTMB(data = donor_spont, Tumors ~ donor * donor_status * receiver, family = binomial,
    REML = F)
m2 <- glmmTMB(data = donor_spont, Tumors ~ donor * donor_status + receiver, family = binomial,
    REML = F)
m3 <- glmmTMB(data = donor_spont, Tumors ~ donor * receiver + donor_status, family = binomial,
    REML = F)
m4 <- glmmTMB(data = donor_spont, Tumors ~ donor + receiver * donor_status, family = binomial,
    REML = F)

m5 <- glmmTMB(data = donor_spont, Tumors ~ donor + donor_status + receiver, family = binomial,
    REML = F)

m6 <- glmmTMB(data = donor_spont, Tumors ~ donor + donor_status, family = binomial,
    REML = F)
m7 <- glmmTMB(data = donor_spont, Tumors ~ donor + receiver, family = binomial, REML = F)
m8 <- glmmTMB(data = donor_spont, Tumors ~ donor_status + receiver, family = binomial,
    REML = F)

m9 <- glmmTMB(data = donor_spont, Tumors ~ donor * donor_status, family = binomial,
    REML = F)
m10 <- glmmTMB(data = donor_spont, Tumors ~ donor * receiver, family = binomial,
    REML = F)
m11 <- glmmTMB(data = donor_spont, Tumors ~ donor_status * receiver, family = binomial,
    REML = F)

m12 <- glmmTMB(data = donor_spont, Tumors ~ donor, family = binomial, REML = F)
m13 <- glmmTMB(data = donor_spont, Tumors ~ donor_status, family = binomial, REML = F)
m14 <- glmmTMB(data = donor_spont, Tumors ~ receiver, family = binomial, REML = F)

m15 <- glmmTMB(data = donor_spont, Tumors ~ 1, family = binomial, REML = F)

AICc(m1, m2, m3, m4, m5, m6, m7, m8, m9, m10, m11, m12, m13, m14, m15)

##     df     AICc
## m1   8 143.6345
## m2   5 143.5008
## m3   5 141.6417
## m4   5 144.6793
## m5   4 142.4959
## m6   3 142.6471
## m7   3 145.5724
## m8   3 140.6781
## m9   4 143.7797
## m10  4 144.5813
## m11  4 142.8254
## m12  2 145.1392
## m13  2 140.8686
## m14  2 143.4691
## m15  1 143.0827

Table of the results of the best fitted models (lower AICc+2)

	Tumors			Tumors			Tumors			Tumors			Tumors
Predictors	Odds Ratios	CI	p	Odds Ratios	CI	p	Odds Ratios	CI	p	Odds Ratios	CI	p	Odds Ratios	CI	p
donor status [T]	2.65	1.09 – 6.46	0.032	2.45	1.03 – 5.86	0.044	2.85	1.11 – 7.31	0.029	2.61	0.73 – 9.33	0.140	2.59	1.06 – 6.33	0.037
receiver [TV]	1.89	0.83 – 4.29	0.131				1.16	0.43 – 3.12	0.770	1.81	0.67 – 4.87	0.242
donor [MT]							0.31	0.08 – 1.25	0.099	0.76	0.31 – 1.86	0.552	0.77	0.32 – 1.85	0.559
donor [MT] × receiver [TV]							4.97	0.80 – 30.98	0.086
receiver [TV] × donor status [T]										1.15	0.19 – 6.91	0.875
Observations	104			104			104			104			104

m13 <- glmmTMB(data = donor_spont, Tumors ~ donor_status, family = binomial, REML = T)
tab_model(m13, show.intercept = F, show.r2 = F, show.re.var = F)

	Tumors
Predictors	Odds Ratios	CI	p
donor status [T]	2.45	1.03 – 5.86	0.044
Observations	104

There is a small but significant effect of the status of the donor, the tumorous donors triggers twice and half more tumors in there grafted host.

Analysis of the tumors development after grafting

Transmissible tumors dataset

Random effects selection

Fixed effects selection

Table of the results of the best fitted models (lower AICc+2)

Spontaneaous tumors dataset

Random effects selection

Fixed effects selection

Table of the results of the best fitted models (lower AICc+2)

Data visualisation