MissingInputException in rule filter_reads_mhcII_SE

[rashidma]

I tested scanneo2 examples and it worked fine. Now, when i run on my data using below command:
snakemake --cores all --configfile /media/user/Drive1/NRC21R-082-03/FFPE/scanneo2/config.yaml --software-deployment-method conda

It gives me error as shown below:

Config file config/config.yaml is extended by additional config specified via the command line.
Assuming unrestricted shared filesystem usage for local execution.
Building DAG of jobs...
MissingInputException in rule filter_reads_mhcII_SE in file /media/user/Drive1/ScanNeo2/workflow/rules/hlatyping.smk, line 301:
Missing input files for rule filter_reads_mhcII_SE:
output: results/BC-FFPE-0003/hla/mhc-II/reads/dna_tumor_DNA_flt_SE.bam
wildcards: sample=BC-FFPE-0003, group=dna_tumor, nartype=DNA
affected files:
results/BC-FFPE-0003/hla/reads/dna_tumor_DNA_SE.fq

Please help to troubleshoot.
Thanks

######################## config.yaml #################################

General Settings

reference:
release: 111
nonchr: false
threads: 30
mapq: 30 # overall required mapping quality
basequal: 20 # overall required base quality

data

data:
name: BC-FFPE-0003
dnaseq:
dna_normal: /media/user/Drive1/NRC21R-082-03/FFPE/rawdata/BC-FFPE-0003-N-exome.fastq
dna_tumor: /media/user/Drive1/NRC21R-082-03/FFPE/rawdata/BC-FFPE-0003-T-exome.fastq
rnaseq:
rna_tumor: /media/user/Drive1/NRC21R-082-03/FFPE/rawdata/BC-FFPE-0003-T-rnaseq.fastq
normal: dna_normal

custom:
variants:
hlatyping:
MHC-I:
MHC-II:

pre-processing (only applied on fastq reads)

preproc:
activate: true # whether (=true) or not (=false) to include pre-processing
minlen: 10
qual: 20
slidingwindow:
activate: true
wsize: 3
wqual: 20

alingment

align:
minovlps: 10
chimsegmin: 20
chimoverhang: 10
chimmax: 50
chimmaxdrop: 30

variant calling

alternative splicing

altsplicing:
activate: true # whether (=true) or not (=false) to include alternative splicing events
confidence: 3 # confidence level (1,2 or 3) - filtering of input alignments
iterations: 5 # number of iteratios (when adding intro edges) - increases sensitivity
edgelimit: 250 # limit max number of edges in graph - affects the runtime

exitron splicing

exitronsplicing:
activate: true # whether (=true) or not (=false) to include exitron-splicing events
ao: 3 # allele observation
pso: 0.05 # percent spliced out
strand: 1 # strand specificity of library (0=unstranded, 1=forward, 2=reverse)

gene fusion

genefusion:
activate: true # whether (=true) or not (=false) to include gene fusion events
maxevalue: 0.3
suppreads: 2 # all fusions with less than suppreads are discarded
maxsuppreads: 1000
maxidentity: 0.3 # genes with fraction of identity are discarded (homologs)
hpolymerlen: 6 # removes breakpoints adjacent to homopolymers of length
readthroughdist: 10000 # distance between breakpoints with less than distance
minanchorlen: 20 # removes fusions whose segments are less than minchimlen
splicedevents: 4 # fusions between genes need at least this many spliced breakpoints
maxkmer: 0.6 # remove reads with repetitive 3-mer that make up more than maxkmer
fraglen: 200 # mean fragment length
maxmismatch: 0.01

indel

indel:
activate: true # whether (=true) or not (=false) to include indels
type: all # long, short, all
mode: BOTH # DNA, RNA or BOTH -

strategy for optimizing posterior probability threshold

strategy: OPTIMAL_F_SCORE # OPTIMAL_F_SCORE, FALSE_DISCOVERY_RATE, CONSTANT
fscorebeta: 1.0 # rel. weight of recall to precision (when OPTIMAL_F_SCORE is selected)
fdr: 0.05 # false discovery rate (when FALSE_DISCOVERY_RATE is selected)
sliplen: 8 # min number of reference bases to suspect slippage event
sliprate: 0.1 # frequency of slippage when it is supsected

quantification:
mode: BOTH # RNA, RNA or BOTH

hlatyping:
class: BOTH # I, II or BOTH

specific path for class II hlatyping (only required when class: II, or BOTH)

MHC-I_mode: DNA # DNA, RNA, or BOTH (if empty alleles have to be specified in custom)
MHC-II_mode: DNA # DNA, RNA, or BOTH (if empty alleles have to be specified in custom)

freqdata: ./hlahd.1.7.1/freq_data/
split: ./hlahd.1.7.1/HLA_gene.split.txt
dict: ./hlahd.1.7.1/dictionary/

prioritization:
class: BOTH # I, II or BOTH
lengths:
MHC-I: 8,9,10,11
MHC-II: 13,14,15

[riasc]

Hi, thanks for the issue.

Its not an optimal solution, but currently, the path of the files defined in the config file needs to be relative to the position from which' snakemake` is executed. Could please test it like this?

As in .tests/integration/custom-test/config/config.yaml where the path is relative to the rootpath of ScanNeo2.

Thanks, I will adjust this for absolute paths later this week

[rashidma]

Hello Richard,
Thanks for the answer. The failed run was using the hlatyping mode = BOTH and prioritization mode = BOTH.
When i changed these modes to type I, it run successfully.

I think scanneo2 considers the files path outside of scanneo2 root directory. please see below:

My scanneo2 root dir : /media/user/Drive1/ScanNeo2
raw data path (mentioned in config file): /media/user/Drive1/NRC21R-082-03/FFPE/rawdata/

And the above is working, great.

I think this was not working due to hlatyping mode =II ?
Please explore this and let me know.
Thanks for your time.

[riasc]

Hi, sorry for the late reply. It seems there was some issue with the processing. Especially when there are no paired-end reads - as HLA-HD requires those. I made some changes to this, and hopefully this should run through now. However, I had to omit the immunogenicity as the webserver of IEDB is not returning the values as it should. I need to think about a solution.

Thanks for letting me know