Limits...
Discovering main genetic interactions with LABNet LAsso-based network inference.

Gadaleta F, Van Steen K - PLoS ONE (2014)

Bottom Line: Despite the valuable results produced thus far, many questions remain unanswered.The permutation-based approach leads to more stable and reliable networks inferred from synthetic microarray data.We show that a higher number of permutations determines the number of predicted edges, improves the overall sensitivity and controls the number of false positives.

View Article: PubMed Central - PubMed

Affiliation: Montefiore Institute University of Liege, Liege, Belgium.

ABSTRACT
Genome-wide association studies can potentially unravel the mechanisms behind complex traits and common genetic diseases. Despite the valuable results produced thus far, many questions remain unanswered. For instance, which specific genetic compounds are linked to the risk of the disease under investigation; what biological mechanism do they act through; or how do they interact with environmental and other external factors? The driving force of computational biology is the constantly growing amount of big data generated by high-throughput technologies. A practical framework that can deal with this abundance of information and that consent to discovering genetic associations and interactions is provided by means of networks. Unfortunately, high dimensionality, the presence of noise and the geometry of data can make the aforementioned problem extremely challenging. We propose a penalised linear regression approach that can deal with the aforementioned issues that affect genetic data. We analyse the gene expression profiles of individuals with a common trait to infer the network structure of interactions among genes. The permutation-based approach leads to more stable and reliable networks inferred from synthetic microarray data. We show that a higher number of permutations determines the number of predicted edges, improves the overall sensitivity and controls the number of false positives.

Show MeSH

Related in: MedlinePlus

True positives and Matthew Correlation Coefficient vs. number of permutations.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4219691&req=5

pone-0110451-g004: True positives and Matthew Correlation Coefficient vs. number of permutations.

Mentions: We also found that the true positive rate follows the same trend of the number of permutations (Figure 4). Within the same figure the Matthew Correlation Coefficient (MCC) is also reported. The MCC is a correlation coefficient between the observed and the predicted classification (presence or absence of edges). It returns a value in the range , where indicates total disagreement between prediction and observations, indicates perfect prediction and no better than random guessing. One important property of the MCC is that it takes into account the number of true negatives and true positives of the predicted network within the normalisation factor. This leads to more meaningful interpretations of the final MCC score. Biological networks are usually sparse. Therefore, prediction methods performed on such networks usually return high numbers of true negatives (absent interactions are correctly predicted). In the extreme case of empty predicted network (a network without any edge), the number of true negatives would positively impact the overall performance of the method. It comes without saying that measuring the number of true negatives would be too optimistic. The MCC mitigates extreme cases of this type. The empty network would have a .


Discovering main genetic interactions with LABNet LAsso-based network inference.

Gadaleta F, Van Steen K - PLoS ONE (2014)

True positives and Matthew Correlation Coefficient vs. number of permutations.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4219691&req=5

pone-0110451-g004: True positives and Matthew Correlation Coefficient vs. number of permutations.
Mentions: We also found that the true positive rate follows the same trend of the number of permutations (Figure 4). Within the same figure the Matthew Correlation Coefficient (MCC) is also reported. The MCC is a correlation coefficient between the observed and the predicted classification (presence or absence of edges). It returns a value in the range , where indicates total disagreement between prediction and observations, indicates perfect prediction and no better than random guessing. One important property of the MCC is that it takes into account the number of true negatives and true positives of the predicted network within the normalisation factor. This leads to more meaningful interpretations of the final MCC score. Biological networks are usually sparse. Therefore, prediction methods performed on such networks usually return high numbers of true negatives (absent interactions are correctly predicted). In the extreme case of empty predicted network (a network without any edge), the number of true negatives would positively impact the overall performance of the method. It comes without saying that measuring the number of true negatives would be too optimistic. The MCC mitigates extreme cases of this type. The empty network would have a .

Bottom Line: Despite the valuable results produced thus far, many questions remain unanswered.The permutation-based approach leads to more stable and reliable networks inferred from synthetic microarray data.We show that a higher number of permutations determines the number of predicted edges, improves the overall sensitivity and controls the number of false positives.

View Article: PubMed Central - PubMed

Affiliation: Montefiore Institute University of Liege, Liege, Belgium.

ABSTRACT
Genome-wide association studies can potentially unravel the mechanisms behind complex traits and common genetic diseases. Despite the valuable results produced thus far, many questions remain unanswered. For instance, which specific genetic compounds are linked to the risk of the disease under investigation; what biological mechanism do they act through; or how do they interact with environmental and other external factors? The driving force of computational biology is the constantly growing amount of big data generated by high-throughput technologies. A practical framework that can deal with this abundance of information and that consent to discovering genetic associations and interactions is provided by means of networks. Unfortunately, high dimensionality, the presence of noise and the geometry of data can make the aforementioned problem extremely challenging. We propose a penalised linear regression approach that can deal with the aforementioned issues that affect genetic data. We analyse the gene expression profiles of individuals with a common trait to infer the network structure of interactions among genes. The permutation-based approach leads to more stable and reliable networks inferred from synthetic microarray data. We show that a higher number of permutations determines the number of predicted edges, improves the overall sensitivity and controls the number of false positives.

Show MeSH
Related in: MedlinePlus