Report Clarification

See allHide authors and affiliations

Science  26 May 2000:
Vol. 288, Issue 5470, pp. 1347
DOI: 10.1126/science.288.5470.1347

This article has a correction. Please see:

In the report “Molecular identification of a eukaryotic, stretch-activated nonselective cation channel” by M. Kanzaki et al. (6 Aug. 1999, p. 882), we mistakenly presented three incorrect traces in Figs. 3 and 4.

In Fig. 3C, the trace at -100 mV was wrong. When we performed this experiment, we saved each trace obtained at various holding potentials in two ways, using Igor Pro (wave matrics). First, we picked up an area (5 s by 10 pA) and pasted it into an area of 13 cm by 7 cm (Series 1). Next, we picked up an area (6 s by 9 pA) and pasted it into an area of 13 cm by 7 cm (Series 2). As a result, the traces of Series 2 were compressed in width and enlarged in amplitude compared with the traces of Series 1. We made the two separate series of data so that we could compare the two series of traces and select the one with better appearance. When making the final version of the traces as an insert for Fig. 3C, we used the traces at 0, -20, -40, -60, and -80 mV from Series 1, but for the -100-mV trace, we incorrectly used the trace at -80 mV from Series 2 instead of the trace at -100 mV of Series 1.

The trace of Fig. 4C was also wrong. During arrangement of the final version of this panel, we placed Fig. 4A instead of the true Fig. 4C. A corrected panel was published earlier (Corrections and Clarifications, 3 Sept. 1999, p. 1493).

In Fig. 4D, we took part of the original control data (which appeared in panel A) by mistake and processed it with a running average method.

The correct traces are shown here. We deeply apologize for having presented the erroneous figures and thank the readers who pointed out the mistakes.

  • Editors' note The above explanations were received after several exchanges with the authors.

Navigate This Article