Figure 2. Linear regression model showing stability of Ebola virus (EBOV) and EBOV RNA in semen at 27°C and 80% relative humidity over 8 days. A) EBOV in bulk (liquid) semen versus dry semen at initial titers of 10⁶ TCID₅₀/mL and 10³ TCID₅₀/mL. The higher titer 1 × 10⁶ TCID₅₀/mL was used to provide a comparison with EBOV in blood, and the lower titer 1 × 10³ TCID₅₀/mL was derived from C_t values reported in semen samples. Viable virus was reduced significantly faster (p<0.0001) in dry semen than in bulk semen. The goodness-of-fit for the linear regression represented as the r² value is 0.53 for bulk semen and 0.82 for dry semen with an initial titer of 10⁶ TCID₅₀/mL, respectively, and 0.65 for bulk semen with an initial titer of 10³ TCID₅₀/mL. No curve is shown for the initial titer 10³ TCID₅₀/mL in the dry semen because no viable virus was recovered after day 1. The titer on day 1 was 1.1 log₁₀ TCID₅₀/mL. In all cases except the high-titer bulk semen sample, the final data point was followed by 2 consecutive days of no recovered virus. B) C_t values produced by analysis of bulk semen samples analyzed by real-time quantitative reverse transcription PCR. The data did not fit a linear regression model (r² = 0.08964), but the RNA clearly remained stable during the entire experiment. Three biologic replicates were analyzed at each time point. Error bars represent mean ± SEM virus titer. Dashed line indicates the limit of detection for the titration assay. An analysis of covariance was used to compare linear regression models and determine differences in virus reduction rates. C_t, cycle threshold; TCID₅₀, 50% tissue culture infectious dose.