Consider a new algorithm for predicting whether a particular RNA binding protein

Question

Consider a new algorithm for predicting whether a particular RNA binding protein binds to an exon. 10,000 exons are evaluated by the prediction method and a cutoff of 2 was selected. Everything scoring above a 2 was considered positive for the RNA binding protein whereas below this level was classified as negative. These results were then compared to a gold standard method of determining whether the RNA binding protein associates with the exon. The results are shown in the following table:

3. Consider a new algorithm for predicting whether a particular RNA binding protein binds to an exon. 10,000 exons are evaluated by the prediction method and a cutoff of 2 was selected. Everything scoring above a 2 was considered positive for the RNA binding protein whereas below this level was classified as negative. These results were then compared to a gold standard method of determining whether the RNA binding protein associates with the exon. The results are shown in the following table Gold Standard" Outcome Prediction Method Positive Negative Total 150 125 Positive 25 375 9850 Negative 9475 Total 500 9500 10,000 Calculate a) Sensitivity b) Specificity c) Positive predictive value

Explanation / Answer

Positive

a)Sensitivity=TP/(TP+FN)= 125/125+25=125/150=0.83=83%

b)Specificity=TN/(TN+FP)=9475/9475+375=9475/9850=0.96=96%

c)Positive predictive value=TP/(TP+FP)=125/125+375=125/500=0.25=25%

Positive

Negative Total Positive 125(True positive-TP) 25 (False Negative-FN) 150 Negative 375(False positive-FP) 9475(True Negative-TN) 9850 Total 500 9500

Related Questions

Navigate

• Browse (All)
• Browse C
• Subjects

---

---

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.