Table 4 A priori estimates of preclinical (pc), clinical (c) and overall (o) drug development success contrasting orthodox (non-genomic) with genomic approaches.

From: Improving the odds of drug development success through human genomics: modelling study

\(\bar{C}\)

γpc

αpc

βpc

FDRpc

Spc

TDRpc = γc

αc

βc

FDRc

TDRc

Sc

So

a

10

0.0001

0.05

0.2

0.9984024

0.05008

0.0015976

0.05

0.2

0.97503657

0.02496343

0.051198203

0.00256

100

0.001

0.05

0.2

0.98423645

0.05075

0.01576355

0.05

0.2

0.79601594

0.20398406

0.06182266

0.00314

1000

0.01

0.05

0.2

0.86086957

0.0575

0.13913043

0.05

0.2

0.27887324

0.72112676

0.154347826

0.00888

10

0.0001

0.00000005

0.2

0.00062455

0.00008

0.99937545

0.05

0.2

0.000039057

0.99996094

0.79953159

0.000064

100

0.001

0.00000005

0.2

0.000062434

0.0008

0.99993757

0.05

0.2

3.9023E-06

0.9999961

0.799953175

0.00064

1000

0.01

0.00000005

0.2

6.1875E-06

0.008

0.99999381

0.05

0.2

3.8672E-07

0.99999961

0.799995359

0.0064

b

10

0.0005

0.05

0.2

0.99205955

0.050375

0.00794045

0.05

0.2

0.8864745

0.1135255

0.055955335

0.00282

100

0.005

0.05

0.2

0.9255814

0.05375

0.074418605

0.05

0.2

0.43736264

0.56263736

0.105813953

0.00569

1000

0.05

0.05

0.2

0.54285714

0.0875

0.45714286

0.05

0.2

0.06909091

0.93090909

0.392857143

0.03438

10

0.0005

0.00000005

0.2

0.00012492

0.00040005

0.99987508

0.05

0.2

7.8085E-06

0.99999219

0.799906309

0.00032

100

0.005

0.00000005

0.2

0.000012437

0.00400005

0.99998756

0.05

0.2

7.7734E-07

0.99999922

0.799990672

0.0032

1000

0.05

0.00000005

0.2

0.000001875

0.04000008

0.99999881

0.05

0.2

7.4219E-08

0.99999993

0.799999109

0.032

  1. TDR, FDR, Spc, Sc and So are presented at different values of α (Type 1 error rate) β (Type 2 error rate) and γ (proportion causal and druggable targets).
  2. (a) \({\gamma }_{pc}=(\bar{C}/{N}_{G})({N}_{T}/{N}_{G})\)when the sample space is defined by \({N}_{G}\times {N}_{D}\), and (b) \({\gamma }_{pc}=(\bar{C}/{N}_{G})({N}_{T}/{N}_{T}\,)\) when the sample space \({N}_{T}\times {N}_{D}\) is restricted to the druggable genome. See text for details.