Limits...
Statistical analysis of a Bayesian classifier based on the expression of miRNAs.

Ricci L, Del Vescovo V, Cantaloni C, Grasso M, Barbareschi M, Denti MA - BMC Bioinformatics (2015)

Bottom Line: We provide a method to enhance a classifiers' performance by exploiting the correlations between the class-discriminating miRNA and the expression of an additional normalized miRNA.By exploiting the normal behavior of triplicate variances and averages, invalid samples (outliers) can be identified by checking their variability via chi-square test or their displacement by the respective population mean via Student's t-test.Finally, the normal behavior allows to optimally set the Bayesian classifier and to determine its performance and the related uncertainty.

View Article: PubMed Central - PubMed

Affiliation: Department of Physics, University of Trento, Trento, I-38123, Italy. leonardo.ricci@unitn.it.

ABSTRACT

Background: During the last decade, many scientific works have concerned the possible use of miRNA levels as diagnostic and prognostic tools for different kinds of cancer. The development of reliable classifiers requires tackling several crucial aspects, some of which have been widely overlooked in the scientific literature: the distribution of the measured miRNA expressions and the statistical uncertainty that affects the parameters that characterize a classifier. In this paper, these topics are analysed in detail by discussing a model problem, i.e. the development of a Bayesian classifier that, on the basis of the expression of miR-205, miR-21 and snRNA U6, discriminates samples into two classes of pulmonary tumors: adenocarcinomas and squamous cell carcinomas.

Results: We proved that the variance of miRNA expression triplicates is well described by a normal distribution and that triplicate averages also follow normal distributions. We provide a method to enhance a classifiers' performance by exploiting the correlations between the class-discriminating miRNA and the expression of an additional normalized miRNA.

Conclusions: By exploiting the normal behavior of triplicate variances and averages, invalid samples (outliers) can be identified by checking their variability via chi-square test or their displacement by the respective population mean via Student's t-test. Finally, the normal behavior allows to optimally set the Bayesian classifier and to determine its performance and the related uncertainty.

No MeSH data available.


Related in: MedlinePlus

Scatter plots of Δx205 (top, left), Δx21 (top, right), −xU6 (bottom). Dots corresponding to samples of the target class ADC and the versus class SQC are colored in blue and red, respectively. In each plot, the black, bold line represents χ, wheres the two black, dashed lines correspond to χ±dχ. Similarly, the three red lines and the three blue lines represent χ10,90, χ10,90±dχ10,90 and χ90,10, χ90,10±dχ90,10, respectively (see Table 4)
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

License 1 - License 2
getmorefigures.php?uid=PMC4559882&req=5

Fig2: Scatter plots of Δx205 (top, left), Δx21 (top, right), −xU6 (bottom). Dots corresponding to samples of the target class ADC and the versus class SQC are colored in blue and red, respectively. In each plot, the black, bold line represents χ, wheres the two black, dashed lines correspond to χ±dχ. Similarly, the three red lines and the three blue lines represent χ10,90, χ10,90±dχ10,90 and χ90,10, χ90,10±dχ90,10, respectively (see Table 4)

Mentions: For each of the three measures of interest, Fig. 2 shows the scatter plot of the respective values as well as the three thresholds χ10:90, χ, χ90:10. The dot color corresponds to the class the sample was assigned to via immunohistochemical analysis and gene profiling (diagnosis). The plots contains four different regions, bounded by the three thresholds and corresponding to different outcomes of the classifier: orange ⇒ versus class with odds larger than 90:10; yellow ⇒ versus class with odds between 50:50 and 90:10; light green ⇒ target class with odds between 50:50 and 90:10; green ⇒ target class with odds larger than 90:10.Fig. 2


Statistical analysis of a Bayesian classifier based on the expression of miRNAs.

Ricci L, Del Vescovo V, Cantaloni C, Grasso M, Barbareschi M, Denti MA - BMC Bioinformatics (2015)

Scatter plots of Δx205 (top, left), Δx21 (top, right), −xU6 (bottom). Dots corresponding to samples of the target class ADC and the versus class SQC are colored in blue and red, respectively. In each plot, the black, bold line represents χ, wheres the two black, dashed lines correspond to χ±dχ. Similarly, the three red lines and the three blue lines represent χ10,90, χ10,90±dχ10,90 and χ90,10, χ90,10±dχ90,10, respectively (see Table 4)
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

License 1 - License 2
Show All Figures
getmorefigures.php?uid=PMC4559882&req=5

Fig2: Scatter plots of Δx205 (top, left), Δx21 (top, right), −xU6 (bottom). Dots corresponding to samples of the target class ADC and the versus class SQC are colored in blue and red, respectively. In each plot, the black, bold line represents χ, wheres the two black, dashed lines correspond to χ±dχ. Similarly, the three red lines and the three blue lines represent χ10,90, χ10,90±dχ10,90 and χ90,10, χ90,10±dχ90,10, respectively (see Table 4)
Mentions: For each of the three measures of interest, Fig. 2 shows the scatter plot of the respective values as well as the three thresholds χ10:90, χ, χ90:10. The dot color corresponds to the class the sample was assigned to via immunohistochemical analysis and gene profiling (diagnosis). The plots contains four different regions, bounded by the three thresholds and corresponding to different outcomes of the classifier: orange ⇒ versus class with odds larger than 90:10; yellow ⇒ versus class with odds between 50:50 and 90:10; light green ⇒ target class with odds between 50:50 and 90:10; green ⇒ target class with odds larger than 90:10.Fig. 2

Bottom Line: We provide a method to enhance a classifiers' performance by exploiting the correlations between the class-discriminating miRNA and the expression of an additional normalized miRNA.By exploiting the normal behavior of triplicate variances and averages, invalid samples (outliers) can be identified by checking their variability via chi-square test or their displacement by the respective population mean via Student's t-test.Finally, the normal behavior allows to optimally set the Bayesian classifier and to determine its performance and the related uncertainty.

View Article: PubMed Central - PubMed

Affiliation: Department of Physics, University of Trento, Trento, I-38123, Italy. leonardo.ricci@unitn.it.

ABSTRACT

Background: During the last decade, many scientific works have concerned the possible use of miRNA levels as diagnostic and prognostic tools for different kinds of cancer. The development of reliable classifiers requires tackling several crucial aspects, some of which have been widely overlooked in the scientific literature: the distribution of the measured miRNA expressions and the statistical uncertainty that affects the parameters that characterize a classifier. In this paper, these topics are analysed in detail by discussing a model problem, i.e. the development of a Bayesian classifier that, on the basis of the expression of miR-205, miR-21 and snRNA U6, discriminates samples into two classes of pulmonary tumors: adenocarcinomas and squamous cell carcinomas.

Results: We proved that the variance of miRNA expression triplicates is well described by a normal distribution and that triplicate averages also follow normal distributions. We provide a method to enhance a classifiers' performance by exploiting the correlations between the class-discriminating miRNA and the expression of an additional normalized miRNA.

Conclusions: By exploiting the normal behavior of triplicate variances and averages, invalid samples (outliers) can be identified by checking their variability via chi-square test or their displacement by the respective population mean via Student's t-test. Finally, the normal behavior allows to optimally set the Bayesian classifier and to determine its performance and the related uncertainty.

No MeSH data available.


Related in: MedlinePlus