<html><head><meta http-equiv="Content-Type" content="text/html charset=utf-8"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class="">Hi, what evidence are you using to get AEDs for the results of your bootstrap training? I don’t find it surprising that the AEDs get worse in subsequent rounds of bootstrap training since overtraining is a real possibility when training ab initio predictors. 300 genes also might not be enough genes, since I think the tutorials and protocols <a href="https://currentprotocols.onlinelibrary.wiley.com/doi/full/10.1002/0471250953.bi0411s48" class="">here</a> and <a href="http://gmod.org/wiki/MAKER_Tutorial_2012#Training_ab_initio_Gene_Predictors" class="">here</a> use 1000 genes for training SNAP. <div class=""><br class=""></div><div class="">I do find it surprising that training file from a different organism gives models that match evidence from your organism of interest. Is that correct? </div><div class=""><br class=""></div><div class="">~Daniel</div><div class=""><div class=""><br class=""></div><div class=""><br class=""><div><blockquote type="cite" class=""><div class="">On Apr 6, 2018, at 9:23 AM, Timo Metz <<a href="mailto:timo.metz@googlemail.com" class="">timo.metz@googlemail.com</a>> wrote:</div><br class="Apple-interchange-newline"><div class=""><div dir="ltr" class=""><div class=""><div class=""><div class=""><div class=""><div class=""><div class=""><div class="">Hello,<br class=""><br class=""></div>I am using MAKER for a non-model organism, and I am currently trying to do the bootstrap training for SNAP as outlined in the tutorial and the paper for MAKER.<br class=""><br class=""></div>For the training I am using a set of ~300 sequences which are conserved (no golden standard genes available) and have very high quality and stop training after third round of bootstrap training. <br class=""><br class=""></div>However, it seems as training does not work properly, because when checking the AEDs for each round of bootstrap training, they actually get worse each round. Also the performance of snap after training is practically similar as before training and significantly worse than using a training file for a model organism.<br class=""><br class=""></div>Are there any suggestions what could be wrong? Is there anything special to check or look at what is not mentioned in the tutorial?<br class=""><br class=""></div>thanks in advance<br class=""><br class=""></div>kind regards<br class=""></div>Timo Metz<br class=""></div>
_______________________________________________<br class="">maker-devel mailing list<br class=""><a href="mailto:maker-devel@box290.bluehost.com" class="">maker-devel@box290.bluehost.com</a><br class="">http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org<br class=""></div></blockquote></div><br class=""></div></div></body></html>