Problem In Cuffdiff Output

Question

[thakkar.bjl]

i have some problem related to cuffidiff output, i used this command for replicates:

cuffdiff -o diff_out -b genome/genome.fa -p 10 -L CT,SS --total-hits-norm -u cuffmerge_experiment/merged.gtf tophat_exp/tophat_exp1/accepted_hits.bam,tophat_exp/tophat_exp2/accepted_hits.bam tophat_exp/tophat_exp3/accepted_hits.bam,tophat_exp/tophat_exp4/accepted_hits.bam

in my output file i found gene column is empty:

TCONS_00000001  XLOC_000001     -       C11044824:403-1007      CT      SS      OK      4.25634 4.05292 -0.0706522      0.148488        0.881958        0.999985        no
TCONS_00000002  XLOC_000002     -       C11047824:1-1042        CT      SS      OK      4.96333 5.54938 0.161018        -0.40853        0.682884        0.999985        no
TCONS_00000003  XLOC_000003     -       C11048020:0-588 CT      SS      OK      3.66925 4.25115 0.212368        -0.422853       0.672403        0.999985        no
TCONS_00000004  XLOC_000004     -       C11048338:216-588       CT      SS      OK      0.795992        0.604691        -0.396557       0.258201        0.796252        0.999985        no
TCONS_00000005  XLOC_000005     -       C11049822:426-1035      CT      SS      OK      0.359691        0.790309        1.13566 -0.675737       0.499208        0.999985        no

i cant understand how can be its possible, if any one have suggestion please guide me.

upto cuffmerge i found gene in my file like this:

C11111420       Cufflinks       exon    726     1304    .       -       .       gene_id "XLOC_000033"; transcript_id "TCONS_00000034"; exon_number "1"; oId "Ca_28192"; nearest_ref "Ca_28192"; class_code "="; tss_id "TSS33"; p_id "P11";
C11112120       Cufflinks       exon    391     1218    .       +       .       gene_id "XLOC_000034"; transcript_id "TCONS_00000035"; exon_number "1"; oId "Ca_27742"; nearest_ref "Ca_27742"; class_code "="; tss_id "TSS34"; p_id "P12";
C11112120       Cufflinks       exon    1577    1657    .       +       .       gene_id "XLOC_000034"; transcript_id "TCONS_00000035"; exon_number "2"; oId "Ca_27742"; nearest_ref "Ca_27742"; class_code "="; tss_id "TSS34"; p_id "P12";
C11112120       Cufflinks       exon    1929    2126    .       -       .       gene_id "XLOC_000035"; transcript_id "TCONS_00000036"; exon_number "1"; oId "Ca_27743"; nearest_ref "Ca_27743"; class_code "="; tss_id "TSS35"; p_id "P13";
C11115550       Cufflinks       exon    6       533     .       +       .       gene_id "XLOC_000036"; transcript_id "TCONS_00000037"; exon_number "1"; oId "Ca_27676"; nearest_ref "Ca_27676"; class_code "="; tss_id "TSS36"; p_id "P14";
C11115550       Cufflinks       exon    2561    2800    .       -       .       gene_id "XLOC_000037"; transcript_id "TCONS_00000038"; exon_number "1"; oId "Ca_27677"; nearest_ref "Ca_2767

so how can be its possible that i didn't find in cuffdiff output. please contact on my email address [e-mail]

[Pierre Lindenbaum]

cross posted on http://seqanswers.com/forums/showthread.php?t=36211

Answer 1

[Fred]

It seems that you don't use a reference GFF/GTF file containing known transcripts and gene names.

You should use a gff/gtf file ( for example the genes.gtf file in one of the archives here http://tophat.cbcb.umd.edu/igenomes.shtml ) in the cuffcompare step:

cuffcompare -r genes.gtf <input1.gtf> [<input2.gtf> .. <inputN.gtf>]}

It produces a cuffcmp.combined.gtf file that you can use in the cuffdiff step.

[thakkar.bjl]

i also used cuffcmp.combined.gtf file in that, i used .gff file then reference fasta file and my cufflink output transcripts.gtf file but it also give same result, in gene column it is like "-" .i tried a lot. thanks for your reply.

[thakkar.bjl]

my gff file look like this.

Ca8     GLEAN   mRNA    37786   42033   0.760266        -       .       ID=Ca_11937;evid_id=GAR_10012294;
Ca8     GLEAN   CDS     41914   42033   .       -       0       Parent=Ca_11937;
Ca8     GLEAN   CDS     41781   41843   .       -       0       Parent=Ca_11937;
Ca8     GLEAN   CDS     41556   41690   .       -       0       Parent=Ca_11937;
Ca8     GLEAN   CDS     41300   41473   .       -       0       Parent=Ca_11937;
Ca8     GLEAN   CDS     41118   41222   .       -       0       Parent=Ca_11937;
Ca8     GLEAN   CDS     40879   40912   .       -       0       Parent=Ca_11937;

[Fred]

Your gtf file seems to be automatically generated by a gene prediction program. So it does not contain any gene annotation (such as gene name, gene id, etc. )