TermTest3_Fall2022_A_updated_small

BIO360H5S page 2 of 18 REFER TO CASE STUDY Study Design Familiarize yourself with the topic (page 11), then refer to the case study description on pages 12 - 13 to answer the questions in this section. Topic: Study design 2 marks 1 Multiple choice: What type of study was this? Select one answer and enter it under question 1 of the Scantron form. (A) Laboratory experiment. (B) Field experiment. (C) Cohort study. (D) Case-control study. (E) Cross-sectional study. 2 marks 2 Multiple choice: Genetic differences between maternal plants are a potential confounding variable. How did the researchers account for it in the study design? Select one answer and enter it under question 2 of the Scantron form. (A) Control. (B) Blocking. (C) Stratification. (D) Blinding. 2 marks 3 Multiple choice: What was the level of replication in this study as it was planned? Select one answer and enter it under question 3 of the Scantron form. (A) 3. (B) 6. (C) 12. (D) 72. 2 marks 4 Multiple choice: If we compared larvae raised on outbred plants from maternal family B1 to larvae raised on inbred plants from the same maternal family, will this result in paired data? Select one answer and enter it under question 4 of the Scantron form. (A) Yes, because the sample size is the same for each breeding type. (B) Yes, because the plants are from the same maternal plant family. (C) No, because the inbred and outbred plants are not the same. (D) No, because each larva is observed for only one breeding type. BIO360H5S page 3 of 18 Topic: Categorical Data Analysis In Box 1 (quoted from Portmann et al. 2015), the researchers described in detail how many individuals were lost due to different reasons. Compare their account to the table of valid sample sizes in Fig. 3. Box 1: a) For the assessment of flight capacity in adult moths (Part 2): "The number of adults available for the flight metabolism assay was significantly reduced because 9 pupae died, 3 adults eclosed with deformed wings, 3 more damaged their wings trying to escape confinement and 1 adult died prematurely. Moths with deformed or damaged wings were excluded from subsequent flight assays." b) For the molecular-level study (Part 3): "RNA isolations for five inbred fed individuals failed, and the reactions were not repeated because the muscle tissue samples had degraded after freeze-thawing." Note: "eclosed" refers to the emergence of adult moths from pupae. Matching: Refer to Box 1 and Fig. 3. For each of the statements listed below, indicate whether it is true (A) or false (B). Use questions 5 – 8 on the Scantron form, one question per statement. 1 mark 5 The loss of valid samples from "Larvae" to "Adults" is consistent with Box 1a. 1 mark 6 The loss of valid samples from "Adults" to "Molecular" is consistent with Box 1b. 1 mark 7 The study design, as it was planned originally, was balanced. 1 mark 8 The study design, based on valid sample sizes for adult moths, is balanced. 2 marks 9 Multiple choice: Use the information in Fig. 3 to calculate the odds ratio for the loss of inbred-fed individuals, from larvae to adults, compared to outbred-fed individuals. What is the correct value of the odds ratio? Select one answer and enter it under question 9 of the Scantron form. (A) 2.5 (B) 1.5. (C) 0.67. (D) 0.4. Matching: Refer to the previous example. For each of the statements listed below, indicate whether it is true (A) or false (B). Use questions 10 – 13 on the Scantron form, one question per statement. 1 mark 10 The hazard ratio may be calculated and interpreted for these data. 1 mark 11 For these data, the odds ratio will provide a good approximation of the hazard ratio. 1 mark 12 The mortality data suggest that outbred plants may be higher quality host plants. 1 mark 13 Differences in mortality rates between groups could introduce bias. BIO360H5S page 4 of 18 Part 1: Growth Differences Refer to pages 14 - 15 to answer the questions in this section. On average, the larvae reared on inbred plants grew faster. This difference translated to the adult stage, where inbred-fed moths had a larger body mass, on average, than outbred-fed moths. Inbred-fed moths also had higher flight muscle mass, on average. Can this difference in flight muscle mass be explained by the difference in body mass alone? Topic: Regression Models Matching: Refer to Fig. 4 and the description of three regression models, A - D, in the table below. Note that the axis does not start at 0. For each of the data (sub)sets listed below, indicate the corresponding model (A, B, C or D). Use questions 14 – 17 on the Scantron form, one question per statement. Model Intercept Slope P-value for slope R 2 Model A 0.274 0.060 0.099 0.070 Model B 0.249 0.101 0.227 0.085 Model C 0.180 0.108 0.017 0.264 Model D 0.318 0.030 0.409 0.019 1 mark 14 Female adult moths. 1 mark 15 Male adult moths. 1 mark 16 All adult moths (not considering whether they are male or female). 1 mark 17 All adult moths, but excluding the two points labeled "P1" and "P2". Topic: Influential Points Matching: Refer to Fig. 4 . Consider the two data points labeled "P1" (a male moth) and "P2" (a female moth) in Fig. 4, and their respective regression models (blue line for point P1, red line for point P2). For each of the statements listed below, indicate whether it is true (A) or false (B). Use questions 18 – 21 on the Scantron form, one question per statement. 1 mark 18 Point P1 has a larger predicted value than point P2. 1 mark 19 Point P 1 has a larger squared residual than point P2. 1 mark 20 If point P1 was excluded, the R 2 of the corresponding model would increase. 1 mark 21 If point P1 was excluded, the slope of the corresponding model would increase. BIO360H5S page 5 of 18 Topic: Residual Analysis The researchers fitted a multiple regression model that combines the effects of two predictors, moth body mass and gender, on thorax mass (response). Box 2 shows how they described the regression results in the published paper. They did not comment on residual analysis, hence your instructor re-analyzed the data in R to check the residuals. Box 2: " Thorax mass (which correlates closely with flight muscle mass [...]) varied with body mass ( P = 0.006) and gender ( P = 0.01), but did not show independent effects of plant breeding type or maternal family ([...] P = 0.009, R 2 = 0.23), suggesting that larval diet had no significant effects on flight muscle mass outside of the effects on overall body size." Notes: "P" refers to a p-value, with separate p-values shown for the two predictors included in the final model (body mass and gender). The R 2 value refers to the combined effect of the two predictors. Plant breeding type and maternal family were not included in the final model. 2 marks 22 Multiple choice: Refer to Box 2 and the residual plots in Fig. 5. What is the appropriate interpretation of these regression results? Select one answer and enter it under question 22 of the Scantron form. (A) Both predictors were statistically significant, and their joint effect was large enough to be practically relevant. (B) Both predictors were statistically significant, but their joint effect was too small to be practically relevant. (C) The predictors were not statistically significant, but there may be a considerable risk of a type II error, more data are needed to be sure. (D) The predictors were not statistically significant, and there was no indication that an important effect may have been missed. (E) The model should not be interpreted, it is not valid because at least one condition has been violated. Topic: Prediction 2 marks 23 Multiple choice: Refer to Fig 6. The dataset is missing the BodyMassg value for one female moth. Interpret the figure visually to decide which of the options listed below is the correct way to estimate the BodyMassg of this moth from its observed value of ThoraxMassg = 0.273? Select one answer and enter it under question 23 of the Scantron form. (A) 1.31. (B) 1.31 ± 0.090. (C) 1.31 ± 0.185. (D) 1.31 ± 0.632. BIO360H5S page 6 of 18 Part 2: Flight Capacity Refer to page 16 to answer the questions in this section. When they analyzed the data on flight capacity, the researchers found that those moths who, as larvae, had been reared on plants from maternal family B1 had a low flight capacity. This suggested that this plant family provided a low quality diet for M. sexta larvae, no matter whether they were inbred-fed or outbred-fed. The researchers worried that this might create bias in their data, hence they reported the results with and without the moths reared on plants from the B1 maternal family. Box 3 quotes the figure caption from the published paper, the corresponding figure is shown in Fig. 7. Box 3: "Peak flight metabolic rate in relation to adult body mass during 5 mins of flight. Open circles and dashed lines represent moths reared on inbred horsenettle plants; solid markers and lines show moths reared on outbred plants. Plot in panel A includes all 3 maternal plant families (inbred: R 2 = 0.50, P < 0.0001, N = 24; outbred: R 2 = 0.12, P = 0.19, N = 16). Family B1, which produced moths with uniformly low metabolic rates [...], is excluded in panel B (inbred: R 2 = 0.32, P = 0.02, N = 16; outbred: R 2 = 0.08, P = 0.39, N = 12)." Topic: Regression Interpretation Matching: Refer to Box 3 and Fig. 7. How did excluding those moths reared on plants from the B1 maternal family affect the regression results for inbred-fed moths? For each of the statements listed below, indicate whether it is true (A) or false (B). Use questions 24 – 27 on the Scantron form, one question per statement. 1 mark 24 Dropping B1 increased the intercept for inbred-fed moths. 1 mark 25 Dropping B1 increased the slope for inbred-fed moths. 1 mark 26 Dropping B1 increased the effect size for inbred-fed moths. 1 mark 27 Dropping B1 increased the p-value for inbred-fed moths. 2 marks 28 Multiple choice: Refer to Box 3 and Fig. 7. Assume that all assumptions and conditions were met. What is the most appropriate biological interpretation of these results? Select the best answer and enter it under question 28 of the Scantron form. (A) Flight capacity (Peak metabolic rate) significantly increased with adult body mass, both for inbred-fed moths and for outbred-fed moths. (B) Flight capacity (Peak metabolic rate) significantly increased with adult body mass for inbred-fed moths but not for outbred-fed moths. (C) Neither inbred-fed not outbred-fed moths showed a significant increase of flight capacity (Peak metabolic rate) with adult body mass. (D) The overall conclusion of the statistical hypothesis tests depended on whether or not the moths raised on plants from the B1 family were excluded. BIO360H5S page 7 of 18 Part 3: Molecular Basis Refer to pages 16 - 17 to answer the questions in this section. The quality of the larval diet may affect flight capacity at a molecular level. The researchers found that differences in flight capacity of adult moths correlated with the relative abundance of troponin t isoform E ( Tnt E ) in the flight muscle. Topic: Hypothesis Testing We learned that different statistical tests are variations on a common theme. Follow the instructions below to perform a one-sided Wilcoxon rank sum test to test whether among the adult male moths that, as larvae, were raised on plants from maternal families B3 or B4, the median ArcSinTntE value is lower for the inbred-fed group (s) than for outbred-fed group (x). The data are shown in Fig. 9. 1. Rank the values of the response variable from the smallest to the largest. 2. For the smaller group (here: s), calculate the sum of the ranks. This will be your test statistic. 3. Compare the test statistic to the critical value for the Wilcoxon test. For the given sample sizes (s: n = 4, x: n = 6), the critical value for a one-sided test, with alternative "less" and significance level alpha = 0.05, is 14. 4. If the test statistic is as small or smaller than the critical value, then the p-value will be smaller than alpha. 5. If the test statistic is larger than the critical value, the p-value will be larger than alpha. 2 marks 29 Multiple choice: After performing the Wilcoxon rank sum test as described above, what can you say about the p-value of this test? Select the best answer and enter it under question 29 of the Scantron form. (A) The p-value will likely be much smaller than 0.05. (B) The p-value will likely be smaller than 0.05, but not by much. (C) The p-value will likely be larger than 0.05, but not by much. (D) The p-value will likely be much larger than 0.05. BIO360H5S page 8 of 18 Topic: Bonferroni Correction The researchers quantified the relative amounts of six different Tnt isoforms (A - F) in the flight muscle sample from each adult moth. They performed a separate ANOVA analysis for each isoform, for a total of six ANOVA models. Each model included two predictors, breeding type and moth gender. They only found a significant effect of breeding type for isoform E. Box 4 quotes how the researchers reported these results. Box 4: "The relative abundance of isoform E was 25.8% lower in inbred fed moths (ANOVA P = 0.011, R 2 = 0.23 breeding type P = 0.049; gender P = 0.008; Table 1), but no independent significant differences in relative abundances were found for the other 5 Tnt isoforms [...]." Notes: "P" refers to a p-value, with separate p-values shown for the overall F-test and for the two predictors breeding type and gender. The R 2 value (= eta squared) refers to the combined effect of the two predictors on the response variable ArcSinTntE. 2 marks 30 Multiple choice: A reviewer for the journal might have requested that the researchers use a Bonferroni correction to account for the number of tests performed on these Tnt data. Refer to Box 4 and apply a Bonferroni correction (with k = 6) to the statistical results reported for isoform E. How would the Bonferroni correction affect the interpretation of the hypothesis tests? Select the best answer and enter it under question 30 of the Scantron form. (A) Both the effect of breeding type and the effect of gender would be statistically significant. (B) Neither the effect of breeding type nor the effect of gender would be statistically significant. (C) The effect of breeding type would be significant, but the effect of gender would not be statistically significant. (D) The effect of gender would be significant, but the effect of breeding type would not be statistically significant. BIO360H5S page 9 of 18 Topic: One-way ANOVA As we have not yet learned how to perform ANOVA with two predictors, your instructor re-analyzed the data using a one-way ANOVA with the four groups shown in Fig. 8, where each group represents a combination of gender and breeding type. The R output is shown in Fig. 10. 2 marks 31 Multiple choice: Refer to Fig. 8. Are the assumptions and conditions for performing an ANOVA on these data met? Select the best answer and enter it under question 31 of the Scantron form. (A) Yes, there is no major issue with the data. (B) No, because the equal means assumption is not met. (C) No, because the equal variances assumption is not met. (D) No, because the data are not independent, they are paired. Matching: Refer to the R output in Fig. 10. Assume that all assumptions and conditions for this ANOVA have been met. For each of the statements listed below, indicate whether it is true (A) or false (B). Use questions 32 – 35 on the Scantron form, one question per statement. 1 mark 32 Based on the data analyzed, the design was balanced. 1 mark 33 The null hypothesis that all means are the same should be rejected. 1 mark 34 Overall, the effect size was large. 1 mark 35 None of the pairwise comparisons was statistically significant. Topic: Strength of Evidence The Carolina horsenettle ( S. carolinense ) is a close relative of cultivated tomato ( S. lycopersicum ), which is a host plant of the tobacco hornworm ( M. sexta ) and its close relative, the tomato hornworm ( M. quinquemaculata ). Matching: Consider the study design and statistical results from this case study. For each of the statements listed below, indicate whether it is true (A) or false (B). Use questions 36 – 39 on the Scantron form, one question per statement. 1 mark 36 The study design allows inferring causality. 1 mark 37 The statistical tests used are all valid. 1 mark 38 The results suggest that moth gender is an important confounding variable when studying the effect of host plant quality on moth flight capacity. 1 mark 39 If the results from this study can be generalized to tomato, commercial seed providers should avoid inbreeding tomato plants, as this may increase herbivory due to increased larval growth and a better ability of adult moths to find tomato plants. Total marks = 50 BIO360H5S page 11 of 18 C ASE STUDY : P LANT I NBREEDING I NCREASES I NSECT F LIGHT C APACITY "You are what you eat" - in the case of the tobacco hornworm ( Manduca sexta ), researchers studied whether small differences in the quality of the host plants, on which the larvae (caterpillars) were reared, affect the flight capacity of adult moths. Earlier research found that inbreeding reduces the ability of the Carolina horsenettle plant ( Solanum carolinense ) to defend itself against herbivores and pathogens, and that tobacco hornworm caterpillars prefer to feed on inbred plants compared to outbred plants. Portman et al. (2017) found that caterpillars that ate inbred plants grew faster and developed into larger pupae compared to caterpillars that ate outbred plants. These differences translated to the adult moth stage: moths that fed on inbred plants as caterpillars exhibited improved flight muscle metabolic function, i.e., they had a better flight capacity. At the molecular level, the differences in flight capacity correlated with changes to the amino acid composition of a key regulatory protein in the flight muscles, troponin t ( Tnt ). Fig. 1 : Life cycle of the tobacco hornworm moth (Manduca sexta; left), and a Manduca sexta larva feeding on a tomato plant (right). Note: The description, analyses and results presented here may differ from the original paper. Reference: • Portman, S.L., Kariyat, R.R., Johnston, M.A., Stephenson, A.G. and Marden, J.H. (2015), Cascading effects of host plant inbreeding on the larval growth, muscle molecular composition, and flight capacity of an adult herbivorous insect. Funct Ecol, 29: 328-337. https://doi-org.myaccess.library.utoronto.ca/10.1111/1365-2435.12358. • Image sources (Fig.1): https://www.rainbowmealworms.net/hornworms, https://commons.wikimedia.org/wiki/File:Tobacco_Hornworm_Manduca_Sexta.jpg