Limits...
Robust dose-response curve estimation applied to high content screening data analysis.

Nguyen TT, Song K, Tsoy Y, Kim JY, Kwon YJ, Kang M, Edberg Hansen MA - Source Code Biol Med (2014)

Bottom Line: The first one is the detection of outliers which is performed during the initialization step with correspondent adjustments of the derivative and error estimation functions.The second aspect is the enhancement of the weighting quality of data points using mean calculation in Tukey's biweight function.Automatic curve fitting of 19,236 dose-response experiments shows that our proposed method outperforms the current fitting methods provided by MATLAB®;'s nlinfit function and GraphPad's Prism software.

View Article: PubMed Central - PubMed

Affiliation: University of California, Davis, USA.

ABSTRACT

Background and method: Successfully automated sigmoidal curve fitting is highly challenging when applied to large data sets. In this paper, we describe a robust algorithm for fitting sigmoid dose-response curves by estimating four parameters (floor, window, shift, and slope), together with the detection of outliers. We propose two improvements over current methods for curve fitting. The first one is the detection of outliers which is performed during the initialization step with correspondent adjustments of the derivative and error estimation functions. The second aspect is the enhancement of the weighting quality of data points using mean calculation in Tukey's biweight function.

Results and conclusion: Automatic curve fitting of 19,236 dose-response experiments shows that our proposed method outperforms the current fitting methods provided by MATLAB®;'s nlinfit function and GraphPad's Prism software.

No MeSH data available.


Two results of outliers and bad fitting.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
getmorefigures.php?uid=PMC4279979&req=5

Fig6: Two results of outliers and bad fitting.

Mentions: Figure 4 shows two examples of fitting when data points do not include outliers and the results of three fitting algorithms are acceptable. Curve fitting results were presented together with sum-of-squares errors (on the top-left corner) which are calculated by taking the sum-of-squares of differences between the actual Y and the estimated Y. Smaller errors indicate better results. The plotting results are shown from left to right: MATLAB®; nlinfit, Prism®; robust fitting, and our method, respectively. Figure 5 illustrates the cases of the presence of outliers. For points inside the interval from −6.5 to −5 (log unit), the variation of measurements is high. In this figure, the first result of MATLAB®; nlinfit demonstrate an ambiguity of the shift parameter: the log of IC50 should be shifted to the right to cross the mean point in the middle of the plot. The first plot of Prism®; presents a poor DRC due to the high steep slope. Figure 6 displays the cases where outliers appear and lead to bad fitting. Prism®; was completely unable to fit the first curve, but our method handled the data points very well. Additionally, the second plot of MATLAB®; nlinfit shows an ambiguity of the shift parameter and a high steep slope.Figure 4


Robust dose-response curve estimation applied to high content screening data analysis.

Nguyen TT, Song K, Tsoy Y, Kim JY, Kwon YJ, Kang M, Edberg Hansen MA - Source Code Biol Med (2014)

Two results of outliers and bad fitting.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
Show All Figures
getmorefigures.php?uid=PMC4279979&req=5

Fig6: Two results of outliers and bad fitting.
Mentions: Figure 4 shows two examples of fitting when data points do not include outliers and the results of three fitting algorithms are acceptable. Curve fitting results were presented together with sum-of-squares errors (on the top-left corner) which are calculated by taking the sum-of-squares of differences between the actual Y and the estimated Y. Smaller errors indicate better results. The plotting results are shown from left to right: MATLAB®; nlinfit, Prism®; robust fitting, and our method, respectively. Figure 5 illustrates the cases of the presence of outliers. For points inside the interval from −6.5 to −5 (log unit), the variation of measurements is high. In this figure, the first result of MATLAB®; nlinfit demonstrate an ambiguity of the shift parameter: the log of IC50 should be shifted to the right to cross the mean point in the middle of the plot. The first plot of Prism®; presents a poor DRC due to the high steep slope. Figure 6 displays the cases where outliers appear and lead to bad fitting. Prism®; was completely unable to fit the first curve, but our method handled the data points very well. Additionally, the second plot of MATLAB®; nlinfit shows an ambiguity of the shift parameter and a high steep slope.Figure 4

Bottom Line: The first one is the detection of outliers which is performed during the initialization step with correspondent adjustments of the derivative and error estimation functions.The second aspect is the enhancement of the weighting quality of data points using mean calculation in Tukey's biweight function.Automatic curve fitting of 19,236 dose-response experiments shows that our proposed method outperforms the current fitting methods provided by MATLAB®;'s nlinfit function and GraphPad's Prism software.

View Article: PubMed Central - PubMed

Affiliation: University of California, Davis, USA.

ABSTRACT

Background and method: Successfully automated sigmoidal curve fitting is highly challenging when applied to large data sets. In this paper, we describe a robust algorithm for fitting sigmoid dose-response curves by estimating four parameters (floor, window, shift, and slope), together with the detection of outliers. We propose two improvements over current methods for curve fitting. The first one is the detection of outliers which is performed during the initialization step with correspondent adjustments of the derivative and error estimation functions. The second aspect is the enhancement of the weighting quality of data points using mean calculation in Tukey's biweight function.

Results and conclusion: Automatic curve fitting of 19,236 dose-response experiments shows that our proposed method outperforms the current fitting methods provided by MATLAB®;'s nlinfit function and GraphPad's Prism software.

No MeSH data available.