[niti217]

I am having similar problem - past 5 hours i spent on debugging error -but in vain.
I would really appreciate any help in this regard.

I am trying to index Homo_sapiens.GRCh37 ...fa file using the command

bwa index -a bwtsw /directory/filename.fa

but it keeps giving me the following error

[bwa_index] Pack FASTA... 56.76 sec
[bwa_index] Reverse the packed sequence... Segmentation fault

Can someone please help me with possible suggestion to fix this. Thank.

[slengyel]

BWA aln error: can't locate index

Greetings,

I'm trying to align my raw paired end illumina reads to my best abyss contigs.fa.

The commands I used to index and align are as follows:

/home/stephen/Programs/BWA/bwa-0.7.3a/bwa index -p contigs.fa -a bwtsw /DATA/ANALYSIS/stephen/k62/contigs.fa

/home/stephen/Programs/BWA/bwa-0.7.3a/bwa aln /DATA/ANALYSIS/stephen/k62/contigs.fa /DATA/RAW_DATA/$1.read1.gz -t 4 >/DATA/ANALYSIS/stephen/$1.read1.sai

I repeat the second command for the read2.gz file.

The indexing appears to go smoothly, i.e. the proper outputs are there. However, when I run the bwa aln command, the following occurs:

[bwa_aln] 17bp reads: max_diff = 2
[bwa_aln] 38bp reads: max_diff = 3
[bwa_aln] 64bp reads: max_diff = 4
[bwa_aln] 93bp reads: max_diff = 5
[bwa_aln] 124bp reads: max_diff = 6
[bwa_aln] 157bp reads: max_diff = 7
[bwa_aln] 190bp reads: max_diff = 8
[bwa_aln] 225bp reads: max_diff = 9
[bwa_aln] fail to locate the index
[main] Version: 0.7.3a-r367

I'm running the commands in the same directory as the index outputs. Why would bwa not be able to find the indices? There seems to be no parameter to tell bwa aln where the index files are located.

The end result is to obtain the metric outputs from CollectInsertSizeMetrics after further picard tools conversions. This is in order to verify insert size and standard deviation values required for input files for ALL-PATHS-LG.

All help is appreciated, and thanks in advance.

[mastal]

BWA & FASTQ or FASTA

Hi,

you don't say how long your Illumina reads are.

bwa aln (BWA-backtrack) only works for reads up to 100 bp, so this could be the problem.

[slengyel]

this particular data set has a read length of 90..two others I was going to attempt later have read lengths of 140.

[mastal]

BWA & FASTQ or FASTA

OK, so it looks like the read length shouldn't be the problem.

Is the bwa index in the same directory as the contigs.fa contigs file? That may be why you get the 'failed to locate the index' error message.

[slengyel]

The bwa index, is not in the same directory as the reference contigs.fa. I'll make a copy of it in the directory where the indices are. I will update with Results.

Thank you for your reply.

[slengyel]

That seems to have done the trick. Thanks again. It's always good to gain a different perspective. Like many, I am new to bioinformatics and assembly.

[rob123king]

BWA index non-recognition issue

Hi,

I have indexed my tomato genome thus:
bwa index -a bwtsw -p S_lycopersicum S_lycopersicum.fa

which does not produce any errors.

reads are 101bp x50 coverage

all files i.e
S_lycopersicum.fa, S_lycopersicum.amb S_lycopersicum.ann S_lycopersicum.bwt S_lycopersicum.pac S_lycopersicum.sa
are located in the same directory but when run this:
bwa mem -pPM -t 10 S_lycopersicum.fa LIB2975_LDI2549_R1L78merged.fastq LIB2975_LDI2549_R2L78merged.fastq > IB2975_LDI2549.sam

I get this error:
[E::bwa_idx_load] fail to locate the index files

Any thoughts on what I'm doing wrong??
Thanks

[ulz_peter]

With the -p parameter you specify the prefix. In the bwa mem command you need to give it the exact same prefix as you specified using the -p argument of the index step:

Hence: drop the .fa suffix of S_lycopersicum.fa

It should look like this
bwa mem -pPM -t 10 S_lycopersicum LIB2975_LDI2549_R1L78merged.fastq LIB2975_LDI2549_R2L78merged.fastq > IB2975_LDI2549.sam

[rob123king]

yes that was it. love ya thanks

[rob123king]

Its running but im getting this at command prompt:
has it not used both files??? does this output look normal?

[W::main_mem] when '-p' is in use, the second query file will be ignored.
[M::main_mem] read 990100 sequences (100000100 bp)...
[M::mem_pestat] # candidate unique pairs for (FF, FR, RF, RR): (7, 1, 5, 5)
[M::mem_pestat] skip orientation FF as there are not enough pairs
[M::mem_pestat] skip orientation FR as there are not enough pairs
[M::mem_pestat] skip orientation RF as there are not enough pairs
[M::mem_pestat] skip orientation RR as there are not enough pairs
[M::worker2@3] performed mate-SW for 0 reads
[M::worker2@7] performed mate-SW for 0 reads
[M::worker2@8] performed mate-SW for 0 reads
[M::worker2@2] performed mate-SW for 0 reads
[M::worker2@9] performed mate-SW for 0 reads
[M::worker2@0] performed mate-SW for 0 reads
[M::worker2@5] performed mate-SW for 0 reads

[ulz_peter]

is there any reason you spcified the -pPM parameters?

if you've got two fastq files skip the -p option. And I#m not sure about the P option; for a first shot give it a try like that:

bwa mem -M -t 10 S_lycopersicum LIB2975_LDI2549_R1L78merged.fastq LIB2975_LDI2549_R2L78merged.fastq > IB2975_LDI2549.sam

[rob123king]

I was looking at the manual and -pP seemed like a good options. -M for picard marking duplicates later. I was just thinking that. Yea I'll give that go, keep it simple.
Thanks

[lh3]

If you have two fastq, one for each end, don't use -p. If you want to perform paired-end alignment, don't use -P. These options are for special use cases. Usually you don't need them. Use -M, though.

[rob123king]

All sorted now thanks very much. completing using samtools mpileup now.