A pivot mutation impedes reverse evolution across an adaptive landscape for drug resistance in Plasmodium vivax

Malaria Journal

Table 1 Summary of simulations of evolution in two schemes

(a) Reverse evolution simulation summary
Starting conditions (drug and allele)	Outcome	Fraction
~3000 μM PYR (peak mutant = 1111)
1111	1111	0.97
1111	1111 → 0111	0.3
~400 μM PYR (peak mutant = 1110)
1111	1111 → 1110	0.93
1111	1111 → 0111	0.06
~55 μM PYR (peak mutant = 1110)
1111	1111 → 1110	0.96
1111	1111 → 0111 → 0110	0.04
~7 μM PYR (peak mutant = 1110)
1111	1111 → 1110	0.97
1111	1111 → 1101	0.03
No drug (peak mutant = 0000)
1111	1111 → 1110	0.91
1111	1111 → 1101	0.08

(b) Double mutants, no drug simulation summary
Outcome	Fraction
Equally distributed: 1100, 1001, 1010, 0011, 0101, 1110
Polymorphic → 1110	0.54
Polymorphic → 0110	0.27
Polymorphic → 1101	0.19

(a) Starting from a population fixed for 1111, across drug concentrations and (b) starting from the centre of the landscape, a population composed equally of the six double mutants. Note that this summary includes several drug concentrations not visualized in Figs. 4 and 5. For (a) which allele in the landscape is the absolute peak at that drug concentration is also highlighted (in parentheses). In a smooth landscape, the landscape should be able to locate the absolute peak. Note that in the no drug environment, the population remains trapped on triple mutant local peaks (1110 and 1101), and unable to locate the 0000 absolute peak

ISSN: 1475-2875