The impact of fitness landscape features on evolutionary outcomes has attracted considerable interest in recent decades. However, evolution often occurs under time-dependent selection in so-called fitness seascapes where the landscape is under flux. Fitness seascapes are an inherent feature of natural environments, where the landscape changes owing both to the intrinsic fitness consequences of previous adaptations and extrinsic changes in selected traits caused by new environments. The complexity of such seascapes may curb the predictability of evolution. However, empirical efforts to test this question utilising a comprehensive set of regimes are lacking. Here we employed an in vitro microbial model system to investigate differences in evolutionary outcomes between time-invariant and -dependent environments, including all possible temporal permutations, with three subinhibitory antimicrobials and a viral parasite (phage) as selective agents. Expectedly, time-invariant environments caused stronger directional selection for resistances compared to time-dependent environments. Intriguingly, however, multidrug resistance outcomes in both cases were largely driven by two strong selective agents (rifampicin and phage) out of four agents in total. These agents either caused cross-resistance or obscured the phenotypic effect of other resistance mutations, modulating the evolutionary outcome overall in time-invariant environments and as a function of exposure epoch in time-dependent environments. This suggests that identifying strong selective agents and their pleiotropic effects is critical for predicting evolution in fitness seascapes, with ramifications for evolutionarily informed strategies to mitigate drug resistance evolution. 

The dataset was collected from a 48-day serial passage experiment through phenotypic measurements at the population or clone level in the lab (curated_pheno_data.txt) or whole-genome sequencing to detect genomic mutations (curated_variants.txt) as specified in the manuscript, metadata, and detailed metadata descriptions below.

curated_variants.txt

"id": unique sample ID (original well plate ID and location of clone as well as antibiotic treatment sequence information)

"plate": 96-well plate ID (1-5 without phage and 1P-5P with phage)

"antibiotic_sequence": antibiotic treatment sequence using understandable abbreviations (no = no antibiotic; nal = nalidixic acid; rif = rifampicin; spe = spectinomycin; all = all antibiotics together throughout experiment)

"strain_orig": original E. coli substrain annotation from sequencing metadata (R6 = 606; R7 = 607-like mutant)

"strain_corrected": E. coli substrain information (R6 = 606; R7 = 607-like mutant) corrected for erroneous annotation in sequencing metadata (this should be used for any analyses on E. coli substrain effect)

"phage": presence of phage (0 = absent; 1 = present)

"chrom": E. coli R606 reference genome accession (NC_012967)  

"pos": nucleotide position on reference genom         

"ref": reference allele 

"alt": alternative (mutated) allele      

"effect": predicted effect of mutated allele (e.g. synonymous_variant or frameshift_variant)         

"effect_simplified": predicted effect of mutated allele simplified into three categories: synonymous, nonsynonymous and intergenic    

"impact": predicted impact of mutated allele (LOW, MODERATE, or HIGH)       

"gene": name of affected gene

"gene_id": ID code of affected gene

"na_change": nucleic acid change in affected gene

"aa_change": amino acid change in affected gene    

"genotype": whether given clone contains given mutation (0 = absence of mutation; 1 = presence of mutation)   

"nal_res": whether given clone is phenotypically resistant to nalidixic acid (0 = susceptible; 1 = resistant)                      

"rif_res": whether given clone is phenotypically resistant to rifampicin (0 = susceptible; 1 = resistant)                 

"spe_res": whether given clone is phenotypically resistant to spectinomycin (0 = susceptible; 1 = resistant)                    

"phage_res": whether given clone is phenotypically resistant to bacteriophage (0 = susceptible; 1 = resistant)     

Missing values and data to be removed prior to analyses are described in detail in the metadata described above.