[E::sam_parse1] CIGAR and query sequence are of different length

[ksahlin]

Hi @marcelm,

I am currently benchmarking v0.7.1 vs 0.8.0 vs commit eb7ef72 for the SNP/indel calling and get the below error

rule make_sorted_bam:
    input: /proj/snic2022-6-31/nobackup/strobealign_eval/alignments_PE/strobealign_eb7ef72/BIO250.sam
    output: /proj/snic2022-6-31/nobackup/strobealign_eval/alignments_PE/strobealign_eb7ef72/BIO250.bam
    jobid: 0
    wildcards: tool=strobealign_eb7ef72, dataset=BIO250

[E::sam_parse1] CIGAR and query sequence are of different length
samtools view: error reading file "/proj/snic2022-6-31/nobackup/strobealign_eval/alignments_PE/strobealign_eb7ef72/BIO250.sam": Input/output error
samtools view: error closing "/proj/snic2022-6-31/nobackup/strobealign_eval/alignments_PE/strobealign_eb7ef72/BIO250.sam": -5
[bam_sort_core] merging from 0 files and 16 in-memory blocks...
[Fri Feb 24 08:27:15 2023]
Error in rule make_sorted_bam:
    jobid: 0
    output: /proj/snic2022-6-31/nobackup/strobealign_eval/alignments_PE/strobealign_eb7ef72/BIO250.bam

RuleException:
CalledProcessError in line 356 of /domus/h1/kris/source/alignment_evaluation/evaluation_VC/Snakefile:
Command 'set -o pipefail;  samtools view -@ 16 -u /proj/snic2022-6-31/nobackup/strobealign_eval/alignments_PE/strobealign_eb7ef72/BIO250.sam | samtools sort -@ 16 -o /proj/snic2022-6-31/nobackup/strobealign_eval/alignments_PE/strobealign_eb7ef72/BIO250.bam' returned non-zero exit status 1.
  File "/proj/snic2020-15-201/anaconda3/envs/scisoseq-eval/lib/python3.9/site-packages/snakemake/executors/__init__.py", line 2318, in run_wrapper
  File "/domus/h1/kris/source/alignment_evaluation/evaluation_VC/Snakefile", line 356, in __rule_make_sorted_bam
  File "/proj/snic2020-15-201/anaconda3/envs/scisoseq-eval/lib/python3.9/site-packages/snakemake/executors/__init__.py", line 560, in _callback
  File "/proj/snic2020-15-201/anaconda3/envs/scisoseq-eval/lib/python3.9/concurrent/futures/thread.py", line 52, in run
  File "/proj/snic2020-15-201/anaconda3/envs/scisoseq-eval/lib/python3.9/site-packages/snakemake/executors/__init__.py", line 546, in cached_or_run
  File "/proj/snic2020-15-201/anaconda3/envs/scisoseq-eval/lib/python3.9/site-packages/snakemake/executors/__init__.py", line 2349, in run_wrapper
Removing output files of failed job make_sorted_bam since they might be corrupted:
/proj/snic2022-6-31/nobackup/strobealign_eval/alignments_PE/strobealign_eb7ef72/BIO250.bam

Here is the run command

strobealign-eb7ef72 -t 16 -r 250 -o /proj/snic2022-6-31/nobackup/strobealign_eval/alignments_PE/strobealign_eb7ef72/BIO250.sam /proj/snic2022-6-31/nobackup/data/HG38/GCA_000001405.15_GRCh38_no_alt_analysis_set.fa /proj/snic2022-6-31/nobackup/data/HG004/BIO250/reads_R1.fastq /proj/snic2022-6-31/nobackup/data/HG004/BIO250/reads_R2.fastq

Full output of the strobealign job below. Also note the warnings (that I have not seen before).

This is strobealign v0.8.0-107-geb7ef72
Time reading reference: 7.53 s
Reference size: 3099.92 Mbp (195 contigs; largest: 248.96 Mbp)
Indexing ...
  Time generating seeds: 186.89 s
  Time estimating number of unique hashes: 10.14 s
  Time sorting non-unique seeds: 2.37 s
  Time generating hash table index: 2.32 s
Total time indexing: 201.78 s
Running in paired-end mode
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Warning: The alignment path of one pair of sequences may miss a small part. [ssw.c ssw_align]
Done!
Total mapping sites tried: 373090331
Total calls to ssw: 40200451
Calls to ksw (rescue mode): 40200451
Did not fit strobe start site: 8738629
Tried rescue: 15625370
Total time mapping: 2146.95 s.
Total time reading read-file(s): 16.42 s.
Total time creating strobemers: 171.01 s.
Total time finding NAMs (non-rescue mode): 343.23 s.
Total time finding NAMs (rescue mode): 165.72 s.
Total time sorting NAMs (candidate sites): 536.19 s.
Total time base level alignment (ssw): 1389.16 s.
Total time writing alignment to files: 0.00 s.
	Command being timed: "strobealign-eb7ef72 -t 16 -r 250 -o /proj/snic2022-6-31/nobackup/strobealign_eval/alignments_PE/strobealign_eb7ef72/BIO250.sam /proj/snic2022-6-31/nobackup/data/HG38/GCA_000001405.15_GRCh38_no_alt_analysis_set.fa /proj/snic2022-6-31/nobackup/data/HG004/BIO250/reads_R1.fastq /proj/snic2022-6-31/nobackup/data/HG004/BIO250/reads_R2.fastq"
	User time (seconds): 33420.12
	System time (seconds): 1108.16
	Percent of CPU this job got: 1463%
	Elapsed (wall clock) time (h:mm:ss or m:ss): 39:19.30
	Average shared text size (kbytes): 0
	Average unshared data size (kbytes): 0
	Average stack size (kbytes): 0
	Average total size (kbytes): 0
	Maximum resident set size (kbytes): 22302784
	Average resident set size (kbytes): 0
	Major (requiring I/O) page faults: 5
	Minor (reclaiming a frame) page faults: 539861156
	Voluntary context switches: 20575
	Involuntary context switches: 60662
	Swaps: 0
	File system inputs: 3451864
	File system outputs: 246435376
	Socket messages sent: 0
	Socket messages received: 0
	Signals delivered: 0
	Page size (bytes): 4096
	Exit status: 0

[ksahlin]

I may also mention that the speed difference between the three versions on this 2*250bp dataset is marginal. Pasted below.

I am starting to believe the v0.7.1 time measures of the NAM sorting is wrong.

Commit eb7ef72 (installed as newest instructions with cmake, i.e. cmake -B build -DCMAKE_C_FLAGS="-march=native" -DCMAKE_CXX_FLAGS="-march=native")

Total mapping sites tried: 373090331
Total calls to ssw: 40200451
Calls to ksw (rescue mode): 40200451
Did not fit strobe start site: 8738629
Tried rescue: 15625370
Total time mapping: 2146.95 s.
Total time reading read-file(s): 16.42 s.
Total time creating strobemers: 171.01 s.
Total time finding NAMs (non-rescue mode): 343.23 s.
Total time finding NAMs (rescue mode): 165.72 s.
Total time sorting NAMs (candidate sites): 536.19 s.
Total time base level alignment (ssw): 1389.16 s.
Total time writing alignment to files: 0.00 s.

v0.8.0 (cmake)

Total mapping sites tried: 373090489
Total calls to ssw: 115247221
Calls to ksw (rescue mode): 40200726
Did not fit strobe start site: 8738600
Tried rescue: 15625370
Total time mapping: 2047.11 s.
Total time reading read-file(s): 16.61 s.
Total time creating strobemers: 181.31 s.
Total time finding NAMs (non-rescue mode): 327.49 s.
Total time finding NAMs (rescue mode): 159.41 s.
Total time sorting NAMs (candidate sites): 512.79 s.
Total time base level alignment (ssw): 1304.86 s.
Total time writing alignment to files: 0.00 s.

v0.7.1

Using rescue cutoff: 84
Running PE mode
Done!
Total mapping sites tried: 374508959
Total calls to ssw: 115870869
Calls to ksw (rescue mode): 40069880
Did not fit strobe start site: 8679805
Tried rescue: 15674982
Total time mapping: 2302.98 s.
Total time reading read-file(s): 20.3734 s.
Total time creating strobemers: 208.319 s.
Total time finding NAMs (non-rescue mode): 428.788 s.
Total time finding NAMs (rescue mode): 186.507 s.
Total time sorting NAMs (candidate sites): 26.6265 s.
Total time reverse compl seq: 0 s.
Total time base level alignment (ssw): 1365.13 s.
Total time writing alignment to files: 0 s.

[ksahlin]

Perhaps related, I also get a segmentation fault error in bcftools mpileup for SIM250 dataset (simulated 2*250bp reads)

rule bcftools_mpileup:
    input: /proj/snic2022-6-31/nobackup/strobealign_eval/alignments_PE/strobealign_eb7ef72/SIM250.bam, /proj/snic2022-6-31/nobackup/data/HG38/GCA_000001405.15_GRCh38_no_alt_analysis_set.fa
    output: /proj/snic2022-6-31/nobackup/strobealign_eval/vcf/strobealign_eb7ef72/SIM250.vcf.gz
    jobid: 0
    wildcards: tool=strobealign_eb7ef72, dataset=SIM250

[mpileup] 1 samples in 1 input files
[mpileup] maximum number of reads per input file set to -d 250
/usr/bin/bash: line 1:  8315 Segmentation fault      (core dumped) bcftools mpileup --threads 16 -O z --fasta-ref /proj/snic2022-6-31/nobackup/data/HG38/GCA_000001405.15_GRCh38_no_alt_analysis_set.fa /proj/snic2022-6-31/nobackup/strobealign_eval/alignments_PE/strobealign_eb7ef72/SIM250.bam > /proj/snic2022-6-31/nobackup/strobealign_eval/vcf/strobealign_eb7ef72/SIM250.vcf.gz
[Fri Feb 24 09:18:51 2023]
Error in rule bcftools_mpileup:
    jobid: 0
    output: /proj/snic2022-6-31/nobackup/strobealign_eval/vcf/strobealign_eb7ef72/SIM250.vcf.gz

RuleException:
CalledProcessError in line 367 of /domus/h1/kris/source/alignment_evaluation/evaluation_VC/Snakefile:
Command 'set -o pipefail;  bcftools mpileup --threads 16 -O z --fasta-ref /proj/snic2022-6-31/nobackup/data/HG38/GCA_000001405.15_GRCh38_no_alt_analysis_set.fa /proj/snic2022-6-31/nobackup/strobealign_eval/alignments_PE/strobealign_eb7ef72/SIM250.bam > /proj/snic2022-6-31/nobackup/strobealign_eval/vcf/strobealign_eb7ef72/SIM250.vcf.gz' returned non-zero exit status 139.

command:

strobealign-eb7ef72 -t 16 -r 250 -o /proj/snic2022-6-31/nobackup/strobealign_eval/alignments_PE/strobealign_eb7ef72/SIM250.sam /proj/snic2022-6-31/nobackup/data/HG38/GCA_000001405.15_GRCh38_no_alt_analysis_set.fa /proj/snic2022-6-31/nobackup/data/HG004/SIM250/reads_R1.fastq /proj/snic2022-6-31/nobackup/data/HG004/SIM250/reads_R2.fastq

Interestingly, On the simulated 2*250bp reads, the time difference in alignments are a lot larger almost 1.4x speedup:

v0.7.1

Total mapping sites tried: 462403523
Total calls to ssw: 49830769
Calls to ksw (rescue mode): 14449902
Did not fit strobe start site: 807582
Tried rescue: 18118441
Total time mapping: 2366.61 s.
Total time reading read-file(s): 31.5706 s.
Total time creating strobemers: 346.908 s.
Total time finding NAMs (non-rescue mode): 723.833 s.
Total time finding NAMs (rescue mode): 216.319 s.
Total time sorting NAMs (candidate sites): 40.3552 s.
Total time reverse compl seq: 0 s.
Total time base level alignment (ssw): 907.439 s.
Total time writing alignment to files: 0 s.

Commit eb7ef72

Total mapping sites tried: 459271867
Total calls to ssw: 14612692
Calls to ksw (rescue mode): 14612692
Did not fit strobe start site: 707189
Tried rescue: 18086605
Total time mapping: 1793.47 s.
Total time reading read-file(s): 28.40 s.
Total time creating strobemers: 289.29 s.
Total time finding NAMs (non-rescue mode): 547.71 s.
Total time finding NAMs (rescue mode): 192.14 s.
Total time sorting NAMs (candidate sites): 779.98 s.
Total time base level alignment (ssw): 643.37 s.
Total time writing alignment to files: 0.00 s.

[ksahlin]

Final note, the cigar error is also present in the BIO150 dataset. Run as:

strobealign-eb7ef72 -t 16 -r 150 -o /proj/snic2022-6-31/nobackup/strobealign_eval/alignments_PE/strobealign_eb7ef72/BIO150.sam /proj/snic2022-6-31/nobackup/data/HG38/GCA_000001405.15_GRCh38_no_alt_analysis_set.fa /proj/snic2022-6-31/nobackup/data/HG004/BIO150/reads_R1.fastq /proj/snic2022-6-31/nobackup/data/HG004/BIO150/reads_R2.fastq

[marcelm]

I was able to reproduce this. Both the warning message ("The alignment path of one pair ...") and the inconsistent CIGAR come from some changes made in SSW. I updated our copy of SSW as part of PR #242. Let me see what can be done about this. Perhaps I should just go back to the older SSW version.

[ksahlin]

Okay great. I guess SSW is still used in the rescue alignment step. Didn't think of that.

[marcelm]

Yes, it’s still used there. I can probably replace it with ksw2 as well.

[marcelm]

Should be fixed now, see my comment in #242.

[ksahlin]

Okay I will start a benchmark of commit edeb390

[ksahlin]

The make_sorted_bam rule completes with the fix in edeb390 for all four datasets (SIM150, SIM250, BIO15,BIO250) but the bcftools_mpileup step still gets a segmentation fault for all four instances.

EXMAPLE COMMAND BCFTOOLS

bcftools mpileup --threads 16 -O z --fasta-ref /proj/snic2022-6-31/nobackup/data/HG38/GCA_000001405.15_GRCh38_no_alt_analysis_set.fa /proj/snic2022-6-31/nobackup/strobealign_eval/alignments_PE/strobealign_main/SIM150.bam > /proj/snic2022-6-31/nobackup/strobealign_eval/vcf/strobealign_main/SIM150.vcf.gz

The four strobealign bam files are found at /proj/snic2022-6-31/nobackup/strobealign_eval/alignments_PE/strobealign_main/*.bam

ERROR:

[mpileup] 1 samples in 1 input files
[mpileup] maximum number of reads per input file set to -d 250
/usr/bin/bash: line 1: 11999 Segmentation fault      (core dumped) bcftools mpileup --threads 16 -O z --fasta-ref /proj/snic2022-6-31/nobackup/data/HG38/GCA_000001405.15_GRCh38_no_alt_analysis_set.fa /proj/snic2022-6-31/nobackup/strobealign_eval/alignments_PE/strobealign_main/SIM150.bam > /proj/snic2022-6-31/nobackup/strobealign_eval/vcf/strobealign_main/SIM150.vcf.gz
[Sat Feb 25 10:13:13 2023]

the error happens after 30s, 1m20s, 1m30s, and 14m in the four files it it is of any help.

[marcelm]

I want to make sure I test this with the same bcftools version as you. Did you just use module load bcftools to make it available?

[marcelm]

I was able to reproduce the problem with bcftools 1.14, 1.15 and 1.16.

[marcelm]

Ah, this is samtools/bcftools#1805, fixed in bcftools 1.17.

I was able to reproduce the problem with a SAM file that looks (shortened) like this:

simulated.265988540	83	chr1	7114818	60	120=1X29=49D	...
simulated.201489102	163	chr1	7114821	60	141=49D9S	...
simulated.294364461	147	chr1	7114822	4	146=49D4=	...

The description in the issue isn’t clear about what the actual problem is, so I can only guess that the long deletion has something to do with it.

Having a CIGAR end with deletions doesn’t make much sense so maybe we should fix that on our side as well. (It does not appear to be the direct cause, though, since there are more alignments with deletions at the end preceding the above records.) It is a result of the NAM being longer on the reference than it should be. Perhaps this is another of these "edge effects", but for the 3' end this time?

[ksahlin]

Perhaps this is another of these "edge effects", but for the 3' end this time?

Yes, probably. At least it matches my intuition. So we have the STR edge effects both in 5' and 3' ends. Since alignments seem to always be left-shifted, this results in an internal deletion in the 5' case. I am wondering why it does not become a softclipp in the 3', and I am in general confused about the cigar string 141=49D9S.

[ksahlin]

I updated to use bcftools v1.17.

• SIM150 failed at a bcftools mpileup step, but it could be an Uppmax related IOError, I will rerun that experiment.
• SIM250 is still running.

Below are the results I currently have comparing v0.7.1, v0.8.0, and commit edeb390 (called main in the table below). Sorry for the current formatting.

Main takaways:

• V0.7.1 and v0.8.0 have identical results (up to a 0.1% precision) in all but two datasets:
  • BIO150 for indels (same recall and precision but F-score diff with <0.1%, probably tipping the rounding decimal)
  • BIO250 for recall and precision in SNV calling - v0.8.0 has higher recall and lower precision.
• New approach (denoted main) has slightly lower SNV recall and similar or slightly lower precision. However, main really lose in the indel prediction with both lower precision and recall (on the biological data at least).

I want to wait until the simulated data is in hand first before making strong conclusions, since the biological indel calling analysis may not be 100% accurate. I simply calculated the intersection (bcftools isec) of indels between gold standard (GIAB) and predictions. SIM datasets should be more controlled. However, it could agree with the intuition that a full semi-global alignment would be able to span some longer indels over the new approach. Let's see when the results from simulated data are ready.

dataset,tool,variant,recall,precision,f_score
BIO150,strobealign_main,SNV,97.8,85.9,91.4
BIO150,strobealign_main,INDEL,7.8,4.7,5.9
BIO150,strobealign_v071,SNV,97.9,85.9,91.5
BIO150,strobealign_v071,INDEL,8.0,5.5,6.6
BIO150,strobealign_v080_cmake,SNV,97.9,85.9,91.5
BIO150,strobealign_v080_cmake,INDEL,8.0,5.5,6.5
-----------------------------------------------
BIO250,strobealign_main,SNV,95.8,81.8,88.3
BIO250,strobealign_main,INDEL,7.6,5.4,6.3
BIO250,strobealign_v071,SNV,95.8,82.0,88.4
BIO250,strobealign_v071,INDEL,7.8,6.2,6.9
BIO250,strobealign_v080_cmake,SNV,95.9,81.9,88.4
BIO250,strobealign_v080_cmake,INDEL,7.8,6.2,6.9
-----------------------------------------------
SIM250,strobealign_main,SNV,98.5,99.5,99.0
SIM250,strobealign_main,INDEL,67.5,59.9,63.5
SIM250,strobealign_v071,SNV,98.5,99.6,99.0
SIM250,strobealign_v071,INDEL,50.6,53.9,52.2
SIM250,strobealign_v080_cmake,SNV,98.5,99.6,99.0
SIM250,strobealign_v080_cmake,INDEL,50.6,53.9,52.2
-----------------------------------------------
SIM150,strobealign_main,SNV,97.3,99.5,98.4
SIM150,strobealign_main,INDEL,64.3,53.6,58.5
SIM150,strobealign_v071,SNV,97.4,99.5,98.4
SIM150,strobealign_v071,INDEL,49.7,53.7,51.6
SIM150,strobealign_v080_cmake,SNV,97.4,99.5,98.4
SIM150,strobealign_v080_cmake,INDEL,49.7,53.7,51.6

Edit: strobealign_main results were added for SIM150 and SIM250 datasets