[adarob]

urchgene,

Why are you converting from BAM? Cufflinks accepted BAM input. My email is adarob at cs.berkeley.edu

[urchgene]

@adarob.........thanks a million for the quick reply

i used the bam as you said..........and sorted with cufflinks recommendation

sort -k 3,3 -k 4,4n tophat_out > tophat_out.sam.sorted

then i run cufflinks and get this error:

cufflinks -N tophat_out.sam.sorted

cufflinks: /usr/lib64/libz.so.1: no version information available (required by cufflinks)
[bam_header_read] EOF marker is absent.
File tophat_out.sam.sorted doesn't appear to be a valid BAM file, trying SAM...
Error: sort order of reads in BAMs must be the same

please, what do i do?

[adarob]

What happens when you just use the bam file output by TopHat as input for Cufflinks?

[urchgene]

@adarob ............when i use output from tophat

the error report is:

[bam_header_read] EOF marker is absent.
File stub_header.sam doesn't appear to be a valid BAM file, trying SAM...
Error: sort order of reads in BAMs must be the same

[adarob]

TopHat now outputs bam, not sam. Are you using an older version of TopHat?

[urchgene]

@adarob ...........The output from tophat ..........


@HD VN:1.0 SO:sorted
@SQ SN:AM168526 LN:577
@SQ SN:AM168527 LN:592
@SQ SN:AM168543 LN:489
@SQ SN:AM168560 LN:529
@SQ SN:AM168607 LN:123
@SQ SN:AM168630 LN:423
@SQ SN:AM168692 LN:492
@SQ SN:AM168702 LN:370
@SQ SN:AM168703 LN:234
@SQ SN:AM168711 LN:592
@SQ SN:AM168727 LN:545
@SQ SN:AM168728 LN:499
@SQ SN:AM168737 LN:450
@SQ SN:AM168757 LN:392
@SQ SN:AM168781 LN:535
@SQ SN:AM168788 LN:376
@SQ SN:AM168798 LN:593
@SQ SN:AM168802 LN:593
@SQ SN:AM168810 LN:566
@SQ SN:AM168818 LN:582
@SQ SN:AM168822 LN:592
@SQ SN:AM168830 LN:198
@SQ SN:AM168859 LN:346
@SQ SN:AM168876 LN:572
@SQ SN:AM168888 LN:456
@SQ SN:AM168905 LN:559
@SQ SN:AM168914 LN:620
@SQ SN:AM168972 LN:645

[caddymob]

Update tophat... I think this was fixed in TopHat 1.1.1 release.

The issue is with the SO tag -- "SO:sorted" needs to be "SO:coordinate"

[caddymob]

@urchgene - you can use Picard ReplaceSamHeader, paste the modified header below into a txt file to fix up the header... I did this before tophat was fixed and it worked for downstream stuff

@HD     VN:1.0  SO:coordinate
@SQ SN:AM168526 LN:577
@SQ SN:AM168527 LN:592
@SQ SN:AM168543 LN:489
@SQ SN:AM168560 LN:529
@SQ SN:AM168607 LN:123
@SQ SN:AM168630 LN:423
@SQ SN:AM168692 LN:492
@SQ SN:AM168702 LN:370
@SQ SN:AM168703 LN:234
@SQ SN:AM168711 LN:592
@SQ SN:AM168727 LN:545
@SQ SN:AM168728 LN:499
@SQ SN:AM168737 LN:450
@SQ SN:AM168757 LN:392
@SQ SN:AM168781 LN:535
@SQ SN:AM168788 LN:376
@SQ SN:AM168798 LN:593
@SQ SN:AM168802 LN:593
@SQ SN:AM168810 LN:566
@SQ SN:AM168818 LN:582
@SQ SN:AM168822 LN:592
@SQ SN:AM168830 LN:198
@SQ SN:AM168859 LN:346
@SQ SN:AM168876 LN:572
@SQ SN:AM168888 LN:456
@SQ SN:AM168905 LN:559
@SQ SN:AM168914 LN:620
@SQ SN:AM168972 LN:645

[jeffhsu3]

This is the error I'm getting:

"Error: this SAM file doesn't appear to be correctly sorted!
current hit is at (null):0, last one was at chrY:59362872
Cufflinks requires that if your file has SQ records in
the SAM header that they appear in the same order as the chromosomes names
in the alignments.
If there are no SQ records in the header, or if the header is missing,
the alignments must be sorted lexicographically by chromsome
name and by position."

This is using cufflinks v0.9.3. Reads were aligned with tophat vn 1.1.2
When I check the read ordering it matches the header file with the exception of the unmapped reads:
chr1
chr10
chr11
chr12
chr13
chr14
chr15
chr16
chr17
chr18
chr19
chr2
chr20
chr21
chr22
chr3
chr4
chr5
chr6
chr7
chr8
chr9
chrM
chrX
chrY
*

There is no entry for the unmapped reads. I tried filtering out the unmapped reads (using samtools view -bf 2 -F 12) from the tophat outpu and keeping only the properly paired reads, but I still got the error.

edit -- "Are you using a GTF annotation? If so, it needs to be in the same order as your SAM file OR you must use a SAM header"
okay getting rid of the GTF annotation solved it. So do I need to make a sam header for the GTF file, or just resort the GTF to the bam file?

edit2 -- spoke too soon. Still gave me that error/

[adarob]

@jeffhsu3,

If you are using the TopHat alignment without modification, this could be a bug in Cufflinks (or TopHat). Check your message for details on how to send me your alignment file to look at. Thanks.

[urchgene]

@adarob..........i am using tophat 1.1.2 as you can see below but i still get the wrong header which gives the error.....

which tophat
/v/linux26_x86_64/appl/molbio/tophat/tophat-1.1.2.Linux_x86_64/tophat


@caddymob...........thanks for your reply but when i use picard ReplaceSamHeader.....

java -Xmx2g -jar ReplaceSamHeader.jar I=stub_header.sam HEADER=header.txt O=Kuusi_2_ACC.bam

..less header.txt

@HD VN:1.0 SO:coordinate

the output file is empty.

[urchgene]

i tried to see what is happening and why the output files are empty. picard-tools runs.

For example i try to convert bowtie aligned reads from sam to fastq

java -Xmx2g -jar SortSam.jar INPUT=solid0179_20100226_Kuusi_2_MCP_pooli_F3 OUTPUT=Kuusi_2_MCP_sorted SORT_ORDER=coordinate

the file solid0179_20100226_Kuusi_2_MCP_pooli_F3 has content like this:

@HD VN:1.0 SO:unsorted
@SQ SN:TC74013 LN:2033
@SQ SN:TC59695 LN:2040
@SQ SN:TC57196 LN:886
@SQ SN:TC57789 LN:859
@SQ SN:TC45664 LN:2103
@SQ SN:TC76604 LN:782
@SQ SN:TC61949 LN:1243
@SQ SN:TC63004 LN:767
@SQ SN:TC46250 LN:1936
@SQ SN:TC84220 LN:1335
@SQ SN:GQ0068.TB_P18 LN:650
@SQ SN:TC82750 LN:1144
@SQ SN:TC83726 LN:1471
@SQ SN:TC77200 LN:1095
@SQ SN:GQ0065.B3_O05 LN:805
@SQ SN:TC82969 LN:1275
@SQ SN:GQ00610.TB.1_N09 LN:854

but the output file Kuusi_2_MCP_sorted is empty.

Please do you have any suggestions or solutions.

[caddymob]

..less header.txt

@HD VN:1.0 SO:coordinate

the output file is empty.

Does your header only have the one @HD line? You need the full header... Seems like picard should error out on this, but maybe not?

You can do "samtools view -H my.bam > header.txt" then modify the header.txt line sorted-->coordinate.

Also on your picard command, did you forget the .bam in your input bam? Also not sure if that matters, but that's all I can see wrong here..

[urchgene]

@caddymob...................thanks, i will try that now.

[urchgene]

@caddymob .......................seems to be there is a problem with this file stub_header.sam

samtools view -H stub_header.sam > header.txt

error:
[bam_header_read] EOF marker is absent.
[bam_header_read] invalid BAM binary header (this is not a BAM file).
[main_samview] fail to read the header from "stub_header.sam".