Extended Data Fig. 7. Oscillations are robust to changes in the species inoculum.

Communities composed of the $\Delta \mathit{\text{tyrA}}$ and $\Delta \mathit{\text{pheA}}$ auxotrophs (n=3) were established at three different inoculum ratios and passaged for a duration of ten days. One community was inoculated at a $\Delta \mathit{\text{tyrA}}:\Delta \mathit{\text{pheA}}$ ratio of 1:9 (left), another at a ratio of 1:1 (middle), and one more at a ratio of 9:1 (right plot). Each plot shows the absolute abundance of the community members measured at the end of each batch. $\Delta \mathit{\text{tyrA}}$ is plotted in cyan while $\Delta \mathit{\text{pheA}}$ is plotted in red. The total community abundance is plotted in grey. Cultures were established and passaged by diluting the cultures to an initial OD₆₀₀ equal to 0.1 at the beginning of each passage. The media was composed of MOPS Minimal Media (Teknova, #M2106) supplemented with 10 μM tyrosine and 20 μM phenylalanine. While similar oscillations emerged in all inoculum conditions, there was a phase difference between the 1:9 inoculum and the 1:1 and 9:1 inocula. This phase difference occurs at the end of the second batch where the community inoculated at 1:9 was dominated by $\Delta \mathit{\text{pheA}}$ while the other communities were dominated by $\Delta \mathit{\text{tyrA}}$. The oscillation phase displayed sensitivity to the concentrations of supplemented amino acids when the inoculum was 1:1. This resulted in phase differences between experiments for a single condition dependent on how the amino acid stock solutions were prepared. To improve reproducibility in our presented data, we selected an inoculum ratio of 9:1 for communities of $\Delta \mathit{\text{tyrA}}$ and $\Delta \mathit{\text{pheA}}$. This always resulted in the same phasing in conditions that produced oscillations.