tag:blogger.com,1999:blog-4376119613549496246.post8574410866419459520..comments2024-03-27T23:36:17.389-07:00Comments on Diving into Genetics and Genomics: log2 fold gene expression change heatmap tommyhttp://www.blogger.com/profile/04023008941349107659noreply@blogger.comBlogger5125tag:blogger.com,1999:blog-4376119613549496246.post-70358090422640479582021-11-01T22:46:34.617-07:002021-11-01T22:46:34.617-07:00All thanks to Mr Anderson for helping with my prof...All thanks to Mr Anderson for helping with my profits and making my fifth withdrawal possible. I'm here to share an amazing life changing opportunity with you. its called Bitcoin / Forex trading options. it is a highly lucrative business which can earn you as much as $2,570 in a week from an initial investment of just $200. I am living proof of this great business opportunity. If anyone is interested in trading on bitcoin or any cryptocurrency and want a successful trade without losing notify Mr Anderson now.Whatsapp: (+447883246472 )<br />Email: tdameritrade077@gmail.comMARGARET MAGOTHEhttps://www.blogger.com/profile/11818424528397171745noreply@blogger.comtag:blogger.com,1999:blog-4376119613549496246.post-72133089599887261582015-06-04T21:07:39.323-07:002015-06-04T21:07:39.323-07:00you may need to specify a smaller height (your max...you may need to specify a smaller height (your max height is 7). say h=3. It all depends on your clustering.<br />ct<- cutree(hc,h=3)tommyhttps://www.blogger.com/profile/04023008941349107659noreply@blogger.comtag:blogger.com,1999:blog-4376119613549496246.post-34719958254524630552015-06-03T22:04:45.057-07:002015-06-03T22:04:45.057-07:00Hi Tommy
I am getting lost in the output of the c...Hi Tommy <br />I am getting lost in the output of the clustering, I have 335 rows and I only get one group after <br /><br />>table(ct) <br /><br />My >plot (hc) has different clusters but maybe the heigh is not fine? The plot shows a maximun height of 7. Don't know how to attach a figure or table in this page to show my example. <br /><br />Thanks,<br /><br />Cata<br /><br /><br /><br /><br />names(hm)<br />m.afterclust<- data[rev(hm$rowInd),rev(hm$colInd)]<br /><br />hc<- as.hclust(hm$rowDendrogram)<br />names(hc)<br />plot(hc) # rotate the dendrogram 90 degree, it is the same as in the heatmap<br /><br />rect.hclust(hc,h=7) # based on the height of the tree, you can specify h<br /><br />ct<- cutree(hc,h=7)<br />table(ct)<br />ct[hc$order]cataguihttps://www.blogger.com/profile/16756845685376074643noreply@blogger.comtag:blogger.com,1999:blog-4376119613549496246.post-30442518704552794502015-04-30T15:07:54.502-07:002015-04-30T15:07:54.502-07:00Sorry, I did not know someone commented on this po...Sorry, I did not know someone commented on this post...<br />I did not use z-score, in the heatmap.2 function, you can specify an argument scale="row" or scale="column", the default is "none".tommyhttps://www.blogger.com/profile/04023008941349107659noreply@blogger.comtag:blogger.com,1999:blog-4376119613549496246.post-69796420289816303622015-01-08T07:15:55.434-08:002015-01-08T07:15:55.434-08:00Hi Tommy,
I have to ask you something, in this he...Hi Tommy,<br /><br />I have to ask you something, in this heatmap are computing the z scores on columns? I have a matrix of up and down regulated genes where I want to generate the heatmap based on log2(FC) of the fragment counts( not gene expression values). So my matrix is a log2 transformed value of ratio of fragments between 2 conditions like tumor vs non tumor having 4 patients. Now my matrix is clearly separated between up and down based on the positive and negative values. I find your code of high value as I want to generate such kind of heatmap for my data. I did produce making some tweaks in this code , only thing I want to understand if this is heatmap zscore is computed or not? and if so is it on columns or not? I feel the z score is not computed, you are taking directly the value of log2 fold change and using pair breaks to get a better visualization of the up and down regulation right? is it important in heatmaps like this that I have to compute the zscore and that too on columns? I am asking since my log2FC values are between tumor vs normal for each row of a gene and has 4 columns where each column represents each patient. Please let me know if you understand my question and if you can reply to my query. Below is the code tweak I did. I get confused between z score and breaks usage. if you can make some clarification of my question. I am clustering the dendrogram on col.<br /><br />data4 is my matrix<br /><br />bk = unique(c(seq(-5.73,-0.5, length=100),seq(-0.5,0, length=100), seq(0,5,length=100)))<br />hmcols<- colorRampPalette(c("green","black", "red"))(length(bk)-1)<br />heatmap.2(data4, breaks=bk, col=hmcols,trace="none", dendrogram = "col",density.info="none",cexCol=0.5, labRow=NA, symm=F,symkey=F,symbreaks=T, scale="none",keysize=1.5)Anonymoushttps://www.blogger.com/profile/01388894543220383691noreply@blogger.com