Limits...
Statistical evaluation of methods for quantifying gene expression by autoradiography in histological sections.

Lazic SE - BMC Neurosci (2009)

Bottom Line: Image segmentation based on thresholding can be subject to floor-effects and lead to biased results.Finally, converting grey level pixel intensities to optical densities or units of radioactivity is unnecessary for most applications and can lead to data with poor statistical properties.Based on these results, suggestions are made to reduce bias, increase precision, and ultimately provide more meaningful results of gene expression data.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Applied Mathematics and Theoretical Physics, Cambridge Computational Biology Institute, University of Cambridge, Cambridge, UK. stan.lazic@cantab.net

ABSTRACT

Background: In situ hybridisation (ISH) combined with autoradiography is a standard method of measuring the amount of gene expression in histological sections, but the methods used to quantify gene expression in the resulting digital images vary greatly between studies and can potentially give conflicting results.

Results: The present study examines commonly used methods for analysing ISH images and demonstrates that these methods are not optimal. Image segmentation based on thresholding can be subject to floor-effects and lead to biased results. In addition, including the area of the structure or region of interest in the calculation of gene expression can lead to a large loss of precision and can also introduce bias. Finally, converting grey level pixel intensities to optical densities or units of radioactivity is unnecessary for most applications and can lead to data with poor statistical properties. A modification of an existing method for selecting the structure or region of interest is introduced which performs better than alternative methods in terms of bias and precision.

Conclusion: Based on these results, suggestions are made to reduce bias, increase precision, and ultimately provide more meaningful results of gene expression data.

Show MeSH
Four methods of selecting the dentate gyrus. The first method used a segmented line drawn down the centre of the DG (B). The background was measured with another line midway between the CA1 and the suprapyramidal blade of the DG. The second method used the polygon tool to outline the DG by hand (C). The third method used a thresholding approach and included only those parts of the DG that were three standard deviations above the mean background GL (D-F). The threshold was adjusted based on the calculated value (D) and then the DG was outlined and the background cleared in order to isolate the DG (E). The outline was only approximate as there was no need to distinguish between background and foreground precisely. The average GL and area were then calculated and the outline of the selected region is shown (F). The fourth method used a mixture model to determine the optimal threshold, and the rest of the analysis was carried out as above. CA = Cornu Ammonis, DG = Dentate gyrus, Hi = Hilus.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2651176&req=5

Figure 1: Four methods of selecting the dentate gyrus. The first method used a segmented line drawn down the centre of the DG (B). The background was measured with another line midway between the CA1 and the suprapyramidal blade of the DG. The second method used the polygon tool to outline the DG by hand (C). The third method used a thresholding approach and included only those parts of the DG that were three standard deviations above the mean background GL (D-F). The threshold was adjusted based on the calculated value (D) and then the DG was outlined and the background cleared in order to isolate the DG (E). The outline was only approximate as there was no need to distinguish between background and foreground precisely. The average GL and area were then calculated and the outline of the selected region is shown (F). The fourth method used a mixture model to determine the optimal threshold, and the rest of the analysis was carried out as above. CA = Cornu Ammonis, DG = Dentate gyrus, Hi = Hilus.

Mentions: A segmented line was drawn down the centre of the DG (Fig. 1B) and the mean grey level was calculated. The background was measured as midway between the CA1 and the suprapyramidal blade of the DG using the same segmented line tool, and both the mean and standard deviation of the background recorded. There is nothing special about using a line, but in the case of the DG, a single line has the property of sampling the majority of the structure while staying away from the edges, which are imprecisely defined. Structures of other shapes can be segmented by 'outlining' them as in Method 2 (below), but staying well within the interior of the structure.


Statistical evaluation of methods for quantifying gene expression by autoradiography in histological sections.

Lazic SE - BMC Neurosci (2009)

Four methods of selecting the dentate gyrus. The first method used a segmented line drawn down the centre of the DG (B). The background was measured with another line midway between the CA1 and the suprapyramidal blade of the DG. The second method used the polygon tool to outline the DG by hand (C). The third method used a thresholding approach and included only those parts of the DG that were three standard deviations above the mean background GL (D-F). The threshold was adjusted based on the calculated value (D) and then the DG was outlined and the background cleared in order to isolate the DG (E). The outline was only approximate as there was no need to distinguish between background and foreground precisely. The average GL and area were then calculated and the outline of the selected region is shown (F). The fourth method used a mixture model to determine the optimal threshold, and the rest of the analysis was carried out as above. CA = Cornu Ammonis, DG = Dentate gyrus, Hi = Hilus.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2651176&req=5

Figure 1: Four methods of selecting the dentate gyrus. The first method used a segmented line drawn down the centre of the DG (B). The background was measured with another line midway between the CA1 and the suprapyramidal blade of the DG. The second method used the polygon tool to outline the DG by hand (C). The third method used a thresholding approach and included only those parts of the DG that were three standard deviations above the mean background GL (D-F). The threshold was adjusted based on the calculated value (D) and then the DG was outlined and the background cleared in order to isolate the DG (E). The outline was only approximate as there was no need to distinguish between background and foreground precisely. The average GL and area were then calculated and the outline of the selected region is shown (F). The fourth method used a mixture model to determine the optimal threshold, and the rest of the analysis was carried out as above. CA = Cornu Ammonis, DG = Dentate gyrus, Hi = Hilus.
Mentions: A segmented line was drawn down the centre of the DG (Fig. 1B) and the mean grey level was calculated. The background was measured as midway between the CA1 and the suprapyramidal blade of the DG using the same segmented line tool, and both the mean and standard deviation of the background recorded. There is nothing special about using a line, but in the case of the DG, a single line has the property of sampling the majority of the structure while staying away from the edges, which are imprecisely defined. Structures of other shapes can be segmented by 'outlining' them as in Method 2 (below), but staying well within the interior of the structure.

Bottom Line: Image segmentation based on thresholding can be subject to floor-effects and lead to biased results.Finally, converting grey level pixel intensities to optical densities or units of radioactivity is unnecessary for most applications and can lead to data with poor statistical properties.Based on these results, suggestions are made to reduce bias, increase precision, and ultimately provide more meaningful results of gene expression data.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Applied Mathematics and Theoretical Physics, Cambridge Computational Biology Institute, University of Cambridge, Cambridge, UK. stan.lazic@cantab.net

ABSTRACT

Background: In situ hybridisation (ISH) combined with autoradiography is a standard method of measuring the amount of gene expression in histological sections, but the methods used to quantify gene expression in the resulting digital images vary greatly between studies and can potentially give conflicting results.

Results: The present study examines commonly used methods for analysing ISH images and demonstrates that these methods are not optimal. Image segmentation based on thresholding can be subject to floor-effects and lead to biased results. In addition, including the area of the structure or region of interest in the calculation of gene expression can lead to a large loss of precision and can also introduce bias. Finally, converting grey level pixel intensities to optical densities or units of radioactivity is unnecessary for most applications and can lead to data with poor statistical properties. A modification of an existing method for selecting the structure or region of interest is introduced which performs better than alternative methods in terms of bias and precision.

Conclusion: Based on these results, suggestions are made to reduce bias, increase precision, and ultimately provide more meaningful results of gene expression data.

Show MeSH