Limits...
Confidence in Phase Definition for Periodicity in Genes Expression Time Series.

El Anbari M, Fadda A, Ptitsyn A - PLoS ONE (2015)

Bottom Line: Separating genes for which we can confidently identify peak activity from ambiguous genes can improve the analysis of time series gene expression.In this study we propose a new statistical method to quantify the phase confidence of circadian genes.The numerical performance of the proposed method has been tested using three real gene expression data sets.

View Article: PubMed Central - PubMed

Affiliation: Division of Biomedical Informatics, Sidra Medical and Research Center, Doha, Qatar.

ABSTRACT
Circadian oscillation in baseline gene expression plays an important role in the regulation of multiple cellular processes. Most of the knowledge of circadian gene expression is based on studies measuring gene expression over time. Our ability to dissect molecular events in time is determined by the sampling frequency of such experiments. However, the real peaks of gene activity can be at any time on or between the time points at which samples are collected. Thus, some genes with a peak activity near the observation point have their phase of oscillation detected with better precision then those which peak between observation time points. Separating genes for which we can confidently identify peak activity from ambiguous genes can improve the analysis of time series gene expression. In this study we propose a new statistical method to quantify the phase confidence of circadian genes. The numerical performance of the proposed method has been tested using three real gene expression data sets.

No MeSH data available.


Graph of the moving block bootstrap principle.Graph showing the principal of moving block bootstrap. The moving block bootstrap randomly selects blocks of the original data (top) and concatenate them together (center) to form a resample (bottom).
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4498625&req=5

pone.0131111.g002: Graph of the moving block bootstrap principle.Graph showing the principal of moving block bootstrap. The moving block bootstrap randomly selects blocks of the original data (top) and concatenate them together (center) to form a resample (bottom).

Mentions: Several bootstrap methods have been proposed for time series data. The most well-known is the Moving Block Bootstrap. This procedure works by dividing the observations in blocks of length b and then resampling the blocks (See Fig 2 for an illustration). The main problem with the block bootstrap is that the block length, b, which is a form of smoothing parameter, needs to be chosen. If the blocks are too short, the bootstrap samples cannot mimic the original sample. In this case dependency is broken whenever we start a new block. If, on the other hand, the blocks are too long, we will lose the randomness of the replicates. For these reasons, in this study we apply the maximum entropy bootstrap algorithm proposed by [8]. It does not impose strong assumptions on the distribution of the time series like stationarity. A full description of the algorithm can be found in [9]. The replications are generated by the following steps


Confidence in Phase Definition for Periodicity in Genes Expression Time Series.

El Anbari M, Fadda A, Ptitsyn A - PLoS ONE (2015)

Graph of the moving block bootstrap principle.Graph showing the principal of moving block bootstrap. The moving block bootstrap randomly selects blocks of the original data (top) and concatenate them together (center) to form a resample (bottom).
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4498625&req=5

pone.0131111.g002: Graph of the moving block bootstrap principle.Graph showing the principal of moving block bootstrap. The moving block bootstrap randomly selects blocks of the original data (top) and concatenate them together (center) to form a resample (bottom).
Mentions: Several bootstrap methods have been proposed for time series data. The most well-known is the Moving Block Bootstrap. This procedure works by dividing the observations in blocks of length b and then resampling the blocks (See Fig 2 for an illustration). The main problem with the block bootstrap is that the block length, b, which is a form of smoothing parameter, needs to be chosen. If the blocks are too short, the bootstrap samples cannot mimic the original sample. In this case dependency is broken whenever we start a new block. If, on the other hand, the blocks are too long, we will lose the randomness of the replicates. For these reasons, in this study we apply the maximum entropy bootstrap algorithm proposed by [8]. It does not impose strong assumptions on the distribution of the time series like stationarity. A full description of the algorithm can be found in [9]. The replications are generated by the following steps

Bottom Line: Separating genes for which we can confidently identify peak activity from ambiguous genes can improve the analysis of time series gene expression.In this study we propose a new statistical method to quantify the phase confidence of circadian genes.The numerical performance of the proposed method has been tested using three real gene expression data sets.

View Article: PubMed Central - PubMed

Affiliation: Division of Biomedical Informatics, Sidra Medical and Research Center, Doha, Qatar.

ABSTRACT
Circadian oscillation in baseline gene expression plays an important role in the regulation of multiple cellular processes. Most of the knowledge of circadian gene expression is based on studies measuring gene expression over time. Our ability to dissect molecular events in time is determined by the sampling frequency of such experiments. However, the real peaks of gene activity can be at any time on or between the time points at which samples are collected. Thus, some genes with a peak activity near the observation point have their phase of oscillation detected with better precision then those which peak between observation time points. Separating genes for which we can confidently identify peak activity from ambiguous genes can improve the analysis of time series gene expression. In this study we propose a new statistical method to quantify the phase confidence of circadian genes. The numerical performance of the proposed method has been tested using three real gene expression data sets.

No MeSH data available.