Can single knockouts accurately single out gene functions?

BMC Systems Biology

Table 1 Results of single vs. multiple perturbations analysis.

	As revealed by single knockouts	As revealed by multiple knockouts
(a) System size	131	215
(b) Cumulative CV of essential genes	96%	94.8%
(c) Mean relative error	45%
(d) Number of essential genes	101

(a) Number of genes implicated as involved in the task. In the multiple case, we only count genes with a statistically significant CV (corrected for multiple testing by FDR [37]: the expected proportion of falsely counted genes is controlled at 10%, or 22 in this case); (b) Total contribution of the essential genes; (c) Mean relative error of CVs; $M R E = \frac{1}{| I |} \cdot \sum_{i \in I} | \frac{γ_{i}^{(1)} - γ_{i}^{(n)}}{γ_{i}^{(n)}} |$ averaging only for the involved genes I; (d) Number of essential genes.

ISSN: 1752-0509