[EMBOSS] db formatting (?) and parsing issue -- emboss version 5.0.0

[George Magklaras]

Indeed there was an \r \n to blame. Didn't spot that with of, because it 
was only one instance at the beginning of the file and not on every 
line. dos2unix to the rescue and we are back in business.

Cheers Peter!

GM

Peter Rice wrote:
> Hi George,
>>  Now, then, we have a second file called 'bhits' (697 sequences). This
>> file has exactly the same format as 'ahits', but when we try to 
>> extract the identified sequences, we get the following:
>>
>> for seq in `cat bhits`; do seqret -filter staphyl68-id:$seq; done
>>
>> Died: seqret terminated: Bad value for '-sequence' with -auto defined
>> 'rror: Unable to read sequence 'staphyl68-id:FLTU7OB01AJHZO
>> (one error per sequence ID)
> 
> Umm ... does the message really start with 'rror'?
> 
> That suggests some non-printing character is involved in the ID. Have 
> you checked bhits does not have any strange characters?
> 
> The error message should be:
> 
> Error: Unable to read sequence 'staphyl68-id:FLTU7OB01AJHZO'
> 
> So something at the end of the ID seems to have moved the final quite to 
> the start of the line.
> 
> I can get the same effect by using noreturn -system pc to change the 
> carriage control characters in bhits.
> 
> I suspect that is the cause of your problem.
> 
> Let me know if that doesn't solve it.
> 
> regards,
> 
> Peter
>