A pattern recognition approach to infer time-lagged genetic interactions

Chuang, Chi‐Cheng; Jen, C.-L.; Chen, Chung-Ming; Shieh, Grace S.

doi:10.1093/bioinformatics/btn098

Cited by 30 publications

(34 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Most algorithms that aim to mine lagged co-clusters do so by working on pairs of rows [11,37,44,51]. They differ in the correlation techniques being used: cross-correlation, normalized, Granger, Pearson, partial, and others [5,16,25, inter alia].…”

Section: Related Workmentioning

confidence: 99%

Sleeved co-clustering of lagged data

2011

View full text Add to dashboard Cite

The paper focuses on mining clusters that are characterized by a lagged relationship between the data objects. We call such clusters lagged co-clusters. A lagged co-cluster of a matrix is a submatrix determined by a subset of rows and their corresponding lag over a subset of columns. Extracting such subsets may reveal an underlying governing regulatory mechanism. Such a regulatory mechanism is quite common in real-life settings. It appears in a variety of fields: meteorology, seismic activity, stock market behavior, neuronal brain activity, river flow, and navigation, but a limited list of examples. Mining such lagged co-clusters not only helps in understanding the relationship between objects in the domain, but assists in forecasting their future behavior. For most interesting variants of this problem, finding an optimal lagged co-cluster is NP-complete problem. We present a polynomial-time Monte-Carlo algorithm for mining lagged co-clusters. We prove that, with fixed probability, the algorithm mines a lagged co-cluster which encompasses the optimal lagged co-cluster by a maximum 2 ratio columns overhead and completely no rows overhead. Moreover, the algorithm handles noise, anti-correlations, missing values, and overlapping patterns. The algorithm is extensively evaluated using both artificial and real-world test environments. The first enable the evaluation of specific, isolated properties of the algorithm. The latter (river flow and topographic data) enable the evaluation of the algorithm to efficiently mine relevant and coherent lagged co-clusters in environments that are temporal, i.e., time reading data and non-temporal.

show abstract

Section: Related Workmentioning

confidence: 99%

Sleeved co-clustering of lagged data

2011

View full text Add to dashboard Cite

show abstract

“…Thus, the threshold of δs could be specified by users based on domain knowledge. [1,2,3,4] are in the same direction as the changes of the values of v ′ for [1,2,3,4], and the changes of the values of v for any two consecutive time points of [7,8,9,10] are in the same direction as the changes of the values of v ′ for [4,5,6,7]. The data movement trends between the time points 4 and 7 in v are not considered due to the gap.…”

Section: Problem Formulationmentioning

confidence: 99%

“…A continuous T pp corresponds to a continuous time segment. For example, {tpp (1,2) , tpp (2,3) , tpp (3,4) } corresponds to time segment {t 1 , t 2 , t 3 , t 4 }. A discontinuous T pp may also corresponds to a continuous time segment.…”

Section: Problem Formulationmentioning

confidence: 99%

“…A discontinuous T pp may also corresponds to a continuous time segment. For example, {tpp (1,2) , tpp (3,4) } correspond to time segment {t 1 , t 2 , t 3 , t 4 }. So, a time segment alone is not sufficient to define the data movements on the time-point pairs and the movement gaps.…”

Section: Problem Formulationmentioning

confidence: 99%

“…Existing methods are not capable of conducting the mining of pan-correlation patterns. They may only be able to detect a special subtype of pan-correlation patterns, for example, positive correlation patterns by [6,3], or negative correlation patterns by [14,7], or both positive and negative correlations by [15,4], or time-lagged positive correlation patterns by [5,2].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Efficient Mining of Pan-Correlation Patterns from Time Course Data

Liu

Wong

et al. 2016

Advanced Data Mining and Applications

View full text Add to dashboard Cite

Abstract. There are different types of correlation patterns between the variables of a time course data set, such as positive correlations, negative correlations, time-lagged correlations, and those correlations containing small interrupted gaps. Usually, these correlations are maintained only on a subset of time points rather than on the whole span of the time points which are traditionally required for correlation definition. As these types of patterns underline different trends of data movement, mining all of them is an important step to gain a broad insight into the dependencies of the variables. In this work, we prove that these diverse types of correlation patterns can be all represented by a generalized form of positive correlation patterns. We also prove a correspondence between positive correlation patterns and sequential patterns. We then present an efficient single-scan algorithm for mining all of these types of correlations. This "pan-correlation" mining algorithm is evaluated on synthetic time course data sets, as well as on yeast cell cycle gene expression data sets. The results indicate that: (i) our mining algorithm has linear time increment in terms of increasing number of variables; (ii) negative correlation patterns are abundant in real-world data sets; and (iii) correlation patterns with time lags and gaps are also abundant. Existing methods have only discovered incomplete forms of many of these patterns, and have missed some important patterns completely.

show abstract

Analysis of Time Course Omics Datasets

Grigorov

2011

Methods in Molecular Biology

View full text Add to dashboard Cite

Over the past 20 years, Omics technologies emerged as the consensual denomination of holistic molecular profiling. These techniques enable parallel measurements of biological -omes, or "all constituents considered collectively", and utilize the latest advancements in transcriptomics, proteomics, metabolomics, imaging, and bioinformatics. The technological accomplishments in increasing the sensitivity and throughput of the analytical devices, the standardization of the protocols and the widespread availability of reagents made the capturing of static molecular portraits of biological systems a routine task. The next generation of time course molecular profiling already allows for extensive molecular snapshots to be taken along the trajectory of time evolution of the investigated biological systems. Such datasets provide the basis for application of the inverse scientific approach. It consists in the inference of scientific hypotheses and theories about the structure and dynamics of the investigated biological system without any a priori knowledge, solely relying on data analysis to unveil the underlying patterns. However, most temporal Omics data still contain a limited number of time points, taken over arbitrary time intervals, through measurements on biological processes shifted in time. The analysis of the resulting short and noisy time series data sets is a challenge. Traditional statistical methods for the study of static Omics datasets are of limited relevance and new methods are required. This chapter discusses such algorithms which enable the application of the inverse analysis approach to short Omics time series.

show abstract

A pattern recognition approach to infer time-lagged genetic interactions

Abstract: Supplementary data and PARE software are available at http://www.stat.sinica.edu.tw/~gshieh/pare.htm.

Cited by 30 publications

References 39 publications

Sleeved co-clustering of lagged data

Sleeved co-clustering of lagged data

Efficient Mining of Pan-Correlation Patterns from Time Course Data

Analysis of Time Course Omics Datasets

Contact Info

Product

Resources

About