differentially expressed genes

Question

[dineshtripathy9658]

The result in my genes_fpkm.tracking file is like

XLOC_000011     XLOC_000011     LOC_Os01g01140     TSS15     Chr1:69674-70131           0         0         0         0         0         0         0
XLOC_000012     XLOC_000012     LOC_Os01g01150     TSS16     Chr1:72774-80256           16.31     12.84     13.42     17.44     14.96     14.89     14.59
XLOC_000013     XLOC_000013     LOC_Os01g01160     TSS17     Chr1:82427-84302           15.5      6.2       12.92     23.95     13.51     20.14     10.82
XLOC_000014     XLOC_000014     LOC_Os01g01170     TSS18     Chr1:85336-88844           1.33      0.32      0.73      1.17      0.67      1.3       0.26
XLOC_000015     XLOC_000015     LOC_Os01g01200     TSS19     Chr1:95498-98558           0         0         0         0         0         0         0

Also from my gene_exp.diff

XLOC_000011     XLOC_000011     LOC_Os01g01140     Chr1:69674-70131     IR64_CT     IR64_DS     0         0         -13.54     0         1
XLOC_000012     XLOC_000012     LOC_Os01g01150     Chr1:72774-80256     IR64_CT     IR64_DS     16.31     12.84     -13.21     -0.44     0.66
XLOC_000013     XLOC_000013     LOC_Os01g01160     Chr1:82427-84302     IR64_CT     IR64_DS     15.5      6.2       -12.68     -1.82     0.24
XLOC_000014     XLOC_000014     LOC_Os01g01170     Chr1:85336-88844     IR64_CT     IR64_DS     1.33      0.32      -12.6      -1.39     0.19
XLOC_000015     XLOC_000015     LOC_Os01g01200     Chr1:95498-98558     IR64_CT     IR64_DS     0         0         -12.56     0         1

I want to find differentially expressed genes. Can I do it from 1st file or I have to go for gene_exp.diff? Also how can a heatmap be generated?

I did make a fpkm matrix out of genes.fpkm_tracking and do I have to take fpkm>=0/1 in any tissue? Also how to calculate up regulated and down regulated?

[Omics data mining]

Hello everyone

I am using cufdiff . I followed steps tophat, -> cufflink->cuffmerge->cuffdiff . I have three query's...

Firstly, I used the gff file in tophat, cufflink and cuffmerge step. For my species, gtf is not available. I want to know using gff instead of gtf will make any difference in result or not?

Secondly I am not getting geneid name in cuffdiff output files.

At last , I lost some of the coordinate in cuffdiff result, initially I had 76000 genes coordinate in my gff file, but I am getting fpkm of 75000 coordinate only in genes.fpkm.tracking

Please reply as soon as possible

Waiting for reply

Answer 1

[GouthamAtla]

May be you should read the documentation once cuffdiff-output-files

Answer 2

[Whoknows]

As @Geek_y says, you can use cuffdiff output file like for DE genes,

For heatmap you can use Pheatmap package in R which helps you to create lots of heatmap so easily.

For up and down regulated genes please see this page: Understanding up and down regulated genes from LOG2 foldchange or foldchange

Answer 3

[Parham]

This workflow can come useful to you!

http://www.bioconductor.org/packages/release/bioc/vignettes/gage/inst/doc/RNA-seqWorkflow.pdf