Replotting MaAsLin3 output in specific column order

Downstream analysis and statistics MaAsLin
1

Hi Will,

I am running MaAsLin3 using the following code:

set.seed(1)
fit_out <- maaslin3(input_data = seqtab.nochim,
                    input_metadata = sample_data,
                    output = "Ox_vs_time",
                    formula = "~ time_point + (1|replicate)",
                    normalization = "TSS",
                    transform = "LOG",
                    augment = TRUE,
                    standardize = TRUE,
                    small_random_effects = TRUE,
                    reference = ("time_point,T1"),
                    max_significance = 0.1,
                    median_comparison_abundance = TRUE,
                    median_comparison_prevalence = FALSE,
                    max_pngs = 25)

and I got the following summary plot:

summary_plot
summary_plot3324×3240 511 KB

I would like to know if it is possible to replot the summary figure so the scatter plot shows first time point 2 and later time point 5? Same with the heatmap, I would like to plot time point 2 first then time point 3, time point 4 and finally time point 5, so I can see the results in a chronological way. Is it possible?

Thanks for your help!
Daniel

2

Yes - if you use coef_plot_vars and heatmap_vars, the order should appear as the vectors you provide to those parameters. Let me know if that doesn’t work like you’re expecting.

Will

3

Hi Will,

Thanks for your prompt reply. Yes, it worked. The code that I ran was:

# Set coef_plot_vars and heatmap_vars
coef_vars <- c('time_point T2', 'time_point T5')
heat_vars <- c('time_point T2', 'time_point T3','time_point T4', 'time_point T5')

# Running MaAsLin3 for Amended samples only
set.seed(1)
fit_out <- maaslin3(input_data = seqtab.nochim,
                    input_metadata = sample_data,
                    output = "Anox_vs_time",
                    formula = "~ time_point + (1|replicate)",
                    normalization = "TSS",
                    transform = "LOG",
                    augment = TRUE,
                    standardize = TRUE,
                    small_random_effects = TRUE,
                    reference = ("time_point,T1"),
                    max_significance = 0.1,
                    median_comparison_abundance = TRUE,
                    median_comparison_prevalence = FALSE,
                    max_pngs = 25,
                    coef_plot_vars = coef_vars,
                    heatmap_vars = heat_vars)

and it was able to generate the desired figure:

summary_plot
summary_plot1920×1791 348 KB

Thanks a lot!

Daniel