| | 1 | = Weighted nearest neighbour imputation, WeNNI = |
| | 2 | |
| | 3 | WeNNI is a method for imputing unreliable expression data |
| | 4 | values. WeNNI imputes the missing values by using continuous spot |
| | 5 | quality weights. In WeNNI the quality of data values are not |
| | 6 | divided into groups of either missing or present, but rather a |
| | 7 | continuous quality weight is assigned to each data value. |
| | 8 | |
| | 9 | The imputation performed by WeNNI is controlled with two parameters: |
| | 10 | the number of nearest neighbours to consider in the calculations and |
| | 11 | beta that is used in the calculation of weights. In the present |
| | 12 | version of WeNNI, the quality weights are calculated from a signal-to-noise |
| | 13 | ratio. The results presented in the WeNNI publication |
| | 14 | show that the choice of parameters is not crucial, and a value around |
| | 15 | 10 as the number of nearest neighbours and a beta in the range 0.1 to |
| | 16 | 1 is suggested when parameter tuning cannot be performed. |
| | 17 | |
| | 18 | WeNNI is described in BMC Bioinformatics where we assessed imputation methods on three data sets containing replicate measurements. Of the compared methods, best performance and robustness were achieved with WeNNI. |
| | 19 | |
| | 20 | ''Improving missing value imputation of microarray data by using spot quality weights'' [[BR]] P. Johansson and J. Häkkinen [[BR]] ''BMC Bioinformatics'' '''7''', 306 (2006) [[BR]] [http://www.biomedcentral.com/1471-2105/7/306/abstract/ Abstract] |
| | 21 | [http://www.biomedcentral.com/1471-2105/7/306 Full Text] |
| | 22 | [http://www.biomedcentral.com/content/pdf/1471-2105-7-306.pdf PDF] |
| | 23 | |
| | 24 | === Comments on WeNNI === |
| | 25 | |
| | 26 | * The notion of weights becomes obsolete after running WeNNI, i.e., do not use the weight fed into WeNNI in any subsequent analysis because all weights are now strictly 1. |
| | 27 | |
| | 28 | * Running WeNNI as a BASE plug-in makes WeNNI destined to impute log ratios of channel 1 and channel 2 (M values in BASE world). A consequence of imputing log ratios is that a change in ratio cannot be assigned to a specific channel. This implies that log(channel1*channel2) (A values in BASE world) become undefined and useless. However, on request from BASE users it was decided that A values should not be affected by the transformation in cases where the A value is well defined before imputation. In cases when an A value do not exist before transformation (i.e. channel1<=0 or channel2<=0) it was decided that A should be set to 0. NOTE, this does not change the underlying WeNNI algorithm in any way but is rather conventions needed for BASE plug-in usage. |
| | 29 | |
| | 30 | = License = |
| | 31 | The WeNNI software is available as a stand alone software package, or |
| | 32 | as a plug-in to BASE, under the [http://www.gnu.org/copyleft/gpl.html GNU General Public License]. |
| | 33 | |
| | 34 | = Download WeNNI = |
| | 35 | |
| | 36 | [attachment:wiki:WeNNI:wenni-0.6.tar.gz?format=raw Download latest stable release (WeNNI-0.6)]. You may need the [attachment:wiki:PluginDownload:BASE-plugindevkit-1.2.17.tar.gz?format=raw BASE 1.2 plug-in development kit]. [[br]] Previous version can be in the ''Attachments'' section below. |
| | 37 | |
| | 38 | === TODO === |
| | 39 | [query:status=new&status=assigned&status=reopened&component=se.lu.thep.wenni&order=priority open tickets] |
| | 40 | [[TicketQuery(status!=closed&component=se.lu.thep.wenni)]] |
