I performed a set of inhibition experiments where Lactobacilli supernatants where tested against a pathogen cell culture. These experiments were carried out in a 96-well plate where pathogen was cultured alone (control), 11 technical replicates, and with a % of lactobacilli supernatant, 3 technical replicates. Absorbance was measured each 2h and growth curves were fitted using grofit R package to calculate growth tax (mu), lag phase (lambda) and max absorbance.

I did an initial screening of +100 lactobacilli strain supernatants using the method described above. In order to perform an initial selection of the best candidate strains I compared statistically using Mann-Whitney U test to compare control technical replicates to each of the strain supernatants cultures contained in a plate ( I had like 21 different supernatants in each plate) since number of replicate were low and normality cannot be tested properly.

Once I got a selection of best inhibiting strains I did 3 biological experiment replicates to get concluding results. Since pathogen does not grow equally among the plates I normalized growth parameters of each 3 biological replicates in a %. I compared all technical replicates from one strain supernatant from the 3 experiments against control technical replicates. I used again Mann-Whitney U test to compare each strain supernatant against a control.

I wonder if Mann-Whitney U is an appropriate statistical analysis in this case because, in fact, I think I am performing a multicomparision since different samples are compared against a control. So, I wonder if using Kruskall-Wallis test and Dunn test as a post-hoc analysis would be the best option. 

Thank you for your advice,


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.