[santiagorevale]

Hi Brian,

Would it be possible to add to bbduk the capability of filtering also the Illumina Index file? Just to clarify what I mean, Illumina produces the following 3 output files for a typical pair-end single indexed sample:

Read 1 : Sample_S1_L001_R1_001.fastq.gz
Read 2 : Sample_S1_L001_R2_001.fastq.gz
Index 1 : Sample_S1_L001_I1_001.fastq.gz

When I'm filtering using bbduk, I specify in1=read1 in2=read2 out1=filtered.read1 out2=filtered.read2.

What I would like is to be able to add "in3=index1 out3=filtered.index1". I don't think any quality filtering should be applied to this file except only outputting only the sequences that passed the quality filtering for reads 1/2.

Currently, what I'm doing now is extracting the IDs of the filtered file and then filtering the index1 file by these IDs, which is very inefficient.

Many new protocols (mainly for single-cell analysis) are using these 3 Illumina files as input in their pipelines. So I think it could be a good addition.

Thank you very much in advance!

Cheers,
S

[bio_d]

aligning Illumina mate-pair library using BBMAP

Code:

bbmap.sh rcomp=t rcs=f in=read1.fq in2=read2.fq out=mapped.sam

Is this the correct way to align illumina mate pair(5.2Kb)/ long mate pair(10Kb) libraries.

No matter what combination of the flags rcs=t/f rcomp=t/f I use the standard error file shows that "Processing reads in paired-ended mode.". I can't understand how to use the flags correctly because in the UsageGuide it is suggested that one should use requirecorrectstrand=f (rcs=f) and rcomp=t for long mate pair. However, since I am getting insert size of 2611.48 when I am expecting something around 5200bp (predicted 5.2Kb Illumina mate pair library from sequencing center) it is most likely the flags are overridden by the default values which are rcs=t and rcomp=f (which is why I presume bbmap is processing reads in the paired-end mode).

Could you please help.

Best,
D

[Gopo]

Hi Gopo,

Yes, I will add that (as an option). Is that common practice in other variant-callers? Note that callvariants.sh does currently have a "PF" (pass filter) field per sample, but I want to make things as simple and compatible as possible.

Hi Brian,
GATK does output "./." for individuals where a genotype can't be reliably called at a particular SNP (can't remember if freebayes and/or ANGSD does this too). What is the flag for "PF" (pass filter) field per sample?

Is there a flag for filtering --min-alleles 2 --max-alleles 2? I ask because I normally do this with VCFtools, but with big VCF files, this takes a while.

Thank you,
Gopo

[GenoMax]

@Brian: K-mer module in FastQC only tracks 2% of the data (1 in 50 sequences)

[Brian Bushnell]

Hello,

I appear to be getting different results when running the same parameters on the same datasets. I used the below commands:

./bbduk.sh in=R1_001.fastq in2=R2_001.fastq minlen=51 mink=11 ktrim=r tbo tpe K=31 hdist=1 hdist2=1 ref=adapters.fa stats=stats_adaptor.txt out=R1_clean.fq out2=R2_clean.fq outm=ER1_adaptormatch.fq outm2=R2_adaptormatch.fq

In the first instance I had a great improvement in the data quality as reported by fastQC with no problems with kmer content in either read. However, when I had to re-run the same data with the same commands, FastQC now flagged up a warning on my R1 reads on the Kmer content. When I ran it a third time it then flagged up a warning on my R2 reads on the Kmer content.

Each time I run the above commands on the dataset I either get:
-No warning on either set of reads for Kmer content
-A warning on R1 Kmer content but not R2
-A warning on R2 Kmer content but not R1

I see two different kmer warnings, either with one or two overrepresented kmers although what kmers and their position in the read differs each time.

Could anyone shed some light on why this is happening? I've attached screenshots of the two-types of Kmer warning that get flagged by fastqc.

Many thanks for your help as always.

Hi Glfrey,

BBDuk is strictly deterministic on a per-read basis, however it does by default reorder the reads when run multithreaded. You can add the flag "ordered" to keep output reads in the same order as input reads, in which case a dif of the output files between multiple runs with the same input and same parameters should indicate that they are identical.

It is possible that FastQC is subsampling or using the first X reads of the file. In either case, the results could differ slightly if BBDuk is not run with the "ordered" flag (both for FastQC and any process that uses a subset of the reads). Please let me know if using the "ordered" flag does not resolve the issue. The number of reads involved here (as reported by FastQC) seems to be very small. Are the output files of the expected size, or did BBDuk perhaps crash for some reason prior to completion?

[Brian Bushnell]

Hi Brian,
I really like callvariants.sh (simulated di-allelic SNPs and at both low (4x) and moderate (60x) sequencing coverage, filtering to keep only variants with quality score of >=Q27, callvariants.sh is more accurate than using GATK 3.7 with their recommended variant filtering). One thing though that I was wondering about, was whether it would be possible to have an option that sets variants that fail for some individuals (in multisample mode) to a missing genotype call "./." rather than whatever genotype is called by callvariants.sh
Best,
Gopo

Hi Gopo,

Yes, I will add that (as an option). Is that common practice in other variant-callers? Note that callvariants.sh does currently have a "PF" (pass filter) field per sample, but I want to make things as simple and compatible as possible.

[glfrey]

Variable Kmer warning when using bbduk

Hello,

I appear to be getting different results when running the same parameters on the same datasets. I used the below commands:

./bbduk.sh in=R1_001.fastq in2=R2_001.fastq minlen=51 mink=11 ktrim=r tbo tpe K=31 hdist=1 hdist2=1 ref=adapters.fa stats=stats_adaptor.txt out=R1_clean.fq out2=R2_clean.fq outm=ER1_adaptormatch.fq outm2=R2_adaptormatch.fq

In the first instance I had a great improvement in the data quality as reported by fastQC with no problems with kmer content in either read. However, when I had to re-run the same data with the same commands, FastQC now flagged up a warning on my R1 reads on the Kmer content. When I ran it a third time it then flagged up a warning on my R2 reads on the Kmer content.

Each time I run the above commands on the dataset I either get:
-No warning on either set of reads for Kmer content
-A warning on R1 Kmer content but not R2
-A warning on R2 Kmer content but not R1

I see two different kmer warnings, either with one or two overrepresented kmers although what kmers and their position in the read differs each time.

Could anyone shed some light on why this is happening? I've attached screenshots of the two-types of Kmer warning that get flagged by fastqc.

Many thanks for your help as always.

[Gopo]

Hi Brian,
I really like callvariants.sh (simulated di-allelic SNPs and at both low (4x) and moderate (60x) sequencing coverage, filtering to keep only variants with quality score of >=Q27, callvariants.sh is more accurate than using GATK 3.7 with their recommended variant filtering). One thing though that I was wondering about, was whether it would be possible to have an option that sets variants that fail for some individuals (in multisample mode) to a missing genotype call "./." rather than whatever genotype is called by callvariants.sh
Best,
Gopo

[dariober]

I don't like to alter sequence headers by default because that could cause the introduction of non-unique headers

Hi Brain- Thanks for the comments (and for the tools of course!). I see your point but I still think the current default behaviour of bbmap is incorrect when sequence names have spaces because it creates an inconsistency between contig names in the sam header versus the names in the sam records.

Since spaces in names are not allowed by the sam format, I would expect bbmap either to silently trim at the first space (which is not nice...!) or fail straightaway with noting that if you really want to proceed you have to enable the tdr=t option. Just a thought...

[Brian Bushnell]

reformat.sh has an option "underscore" which will change whitespace in sequence headers into underscores, if the extra information is important. Alternatively, as Genomax says, you can use "trimrname". Generally I don't like to alter sequence headers by default because that could cause the introduction of non-unique headers, which would break all tools, rather than just some tools

[GenoMax]

You can choose to truncate the names at the first space during mapping using "trd=t".

Since you already have sam/bam files you can use reformat.sh to take care of the spaces using

trimrname=f             For sam/bam files, trim rname/rnext fields after the first space.

[dariober]

Sorry... another issue.

I have a reference genome where the header lines contain spaces, like this:

>chr1  AC:CM000663.2  gi:568336023  LN:248956422  rl:Chromosome  M5:6aef897c3d6ff0c78aff06ac189178dd  AS:GRCh38
NNNNN...

After mapping with bbmap I get a sam header where the SN field contains the full fasta header:

cat -vet test.sam | head

@HD^IVN:1.4^ISO:unsorted$
@SQ^ISN:chr1  AC:CM000663.2  gi:568336023  LN:248956422  rl:Chromosome  M5:6aef897c3d6ff0c78aff06ac189178dd  AS:GRCh38^ILN:248956422$

the sam records contain only the first word, i.e. "chr1", "chr2" etc. This causes problems to downstream tools since "chr1 AC:CM000663.2 ..." is not a valid reference name.

I think the printing of the sequence name in the sam header should strip everything after the first blank space?

[dariober]

Hi Dario and Genomax,

I though I had gotten rid of that a while ago, but I guess not - I'll investigate and fix it. It's harmless and due to a race condition for a thread finishing after it was prematurely shut down because enough reads were generated, but it looks scary.

Hi Brain, thanks for feedback. I thought it was harmless the problem is that I have a pipeline (snakemake) that will "refuse" to proceed on receiving this error.

[Brian Bushnell]

1) I have not looked at Nanopore reads in several years now so I don't have any further experience in mapping them. But from what I read and hear the quality is gradually creeping upward (and my initial exposure was to very low quality data) so some of the settings can probably be relaxed; e.g., increasing to k=10 and minratio=0.25.

2) "Read Depth" (DP field) is the depth of all reads covering (or spanning) the event, while "Allele Depth" (AD field) is the same but only for those reads exhibiting the variation.

Coverage is added to an array tracking per-reference-base coverage at each position when each read is processed (and deleted bases are considered covered by a read containing that deletion). Similarly, coverage is added to each variation when a read containing that variation is processed.
The reference coverage for a variant is calculated by summing the coverage per-base in the array over the length of the event. For substitutions this is straightforward. For deletions, this is the deleted reference bases. For insertions, this is the average coverage of the bases on either side of the insertion.
It is possible for the allele coverage to be slightly higher than the total coverage using this calculation method (for example, as a result of a sudden drop in coverage on one side of an insertion event) so DP reports max(ref coverage, allele coverage) while COV reports the raw calculated coverage (they will virtually always be the same, though). AF is calculated as DP/AD to prevent AF over 1.0.

3) "Allele Fraction" is "Allele Depth"/"Read Depth". As you say, this is straightforward for SNPs but less so for indels, particularly with short reads and long deletions. It's not entirely clear what AF should really mean for a deletion, because it describes a ratio of two classes of reads - those that cover a variant location and do not indicate the variant, and those that cover a variant location and do indicate the variant. But when deletions are longer than reads there is a third case; those that land in the middle of the deletion but do not touch either end. You could, for example, have 100x coverage on either end of a deletion event, with 50x indicating the deletion and 50x indicating no deletion, and calculate an AF of 0.5. But if you had 1000x coverage somewhere in the middle of the deletion event from reads not spanning the junction on either end, that's a 3rd category and really they should not contribute to the AF (since the deleted allele would never be observed in that case), but currently they do. I'll look into changing the way that long deletion AF and DP are calculated; thanks for making me think about it.
For insertions, there is "Alternate Allele Frequency" which is allele frequency adjusted to compensate for the fact that insertions will not be called on reads that do not span the full event. So if you have a 50bp insertion and 100bp reads, only 1/3rd of the reads with bases in the inserted region would actually span the full insertion and thus possibly yield an insertion call, so a 100% true allele frequency might be reported as 33%. "Alternate Allele Frequency" would adjust the 33% observed back up to 100% based on the relative average read length and insertion length.

[jweger1988]

Hi Brian,

I have a couple of questions about using callvariants.sh and mapPacbio.sh. I am using mapPacbio.sh to align reads from a Nanopore minion to a viral sequence and looking for deletions with callvariants.sh. My questions are

1. Do you have a suggested method for mapping minion reads. I used a suggestion you provided in 2014 but I don't know if you have since updated this. Here is what I'm running.

mapPacBio.sh k=8 in=reads.fq ref=ref.fa maxreadlen=1000 minlen=200 sam=1.4 idtag ow int=f qin=33 indelhist=indelhist1.txt out=mapped.sam minratio=0.15 ignorequality slow ordered maxindel=12000 outputunmapped=f bs=bs1.sh

2. I am then calling variants with the desire to look for deletions with callvariants.sh.

callvariants.sh in=mapped.sam rarity=0.0001 out=deletions.tsv vcf=deletions.vcf ref=ref.fa extended=t border=20 minscore=10

In the output can you tell me what the meaning of "coverage" or "read Depth" or "allele depth" are? What is the difference between these? Also, what does the allele fraction mean in this case? For a SNP it's self-explanatory but for a indel it seems like it's not as straight forward.

Any insight you can provide would be extremely useful. Thanks again for all you do.