[Tally]

When importing sam to bam, I use 'view' rather than 'import' :

@SQ header lines present in SAM file:
samtools view –bS alignment.sam > alignment.bam

@SQ header lines absent from SAM file:
samtools view –bt reference.fasta.fai alignment.sam > alignment.bam

Not sure if that will help at all but might be worth a try
Good luck

[HeidiJTP]

Thanks for the suggestion Tally, I just reran everything and used the command view instead of import, but still no luck

[HeidiJTP]

I think there are two problems here:

1) Display of NNNNs instead of sequence
This seems to be related in part to the actual terminal window. I thought it was weird that the NNNs don't appear until exactly after I start scrolling across the terminal. If I resize the terminal before running the 'tview' command, the position where the NNNs begin also changes. It may not be important, as according to mpileup output the NNNs are only occurring in between aligned regions.

2) Incomplete alignment
My fault!!! Helps when you use the correct reference sequence...

[sudeep]

I think there are two problems here:

1) Display of NNNNs instead of sequence
This seems to be related in part to the actual terminal window. I thought it was weird that the NNNs don't appear until exactly after I start scrolling across the terminal. If I resize the terminal before running the 'tview' command, the position where the NNNs begin also changes. It may not be important, as according to mpileup output the NNNs are only occurring in between aligned regions.

I am also facing this problem. Does anybody know what is the work around ?

[Crypticfortune]

I am also facing this problem. Does anybody know what is the work around?

To summarize, I think there's 3 main problems that trip up users:

1. Forgot to specify the reference on the command line (eg. "samtools tview foo.bam" => "samtool tview foo.bam foo.fa")
2. Fasta file has different names for sequences. This is painful to fix, but you'll have to either rewrite all the sequence names (e.g. ">chr1" lines in foo.fa) to match the bam file sequence names, or rewrite the sequence references in the SAM/BAM file. The former's probably easier, but definitely the "right" way to go is to use the same fasta files when building the alignment to begin with
3. Corrupt fasta files? I can't confirm this, but I suspect samtools might choke on reading FASTA files with dos/windows CR/LF linebreak codes (shows up as ^M in unix terminals a lot). This would explain HeidiJTP and naluru's 80 character problem (as 80 characters per line is common). You can normalize your dos/windows ASCII files to unix with the dos2unix command (e.g. dos2unix foo.fa).

Also, it may not be what you're looking for if you care about the reference outside of mapped areas, but as an alternative, Samscope infers and displays the reference from BAM data alone (MD + CIGAR tags) without relying on FASTA reference files.

[kristina.gagalova]

I have solved this issue with renaming the fai (fasta index file). I had FileName.fa.fai and rename it FileName.fai. I think the program expects it like that.