Compare Three Annotators For Genetic Interpretation

---

ClinVar Data Collection and Curation

You can download our normalized VCF file and decompress it for annotation:

• ClinVar_PLP
• ClinVar_BLB

Or, See ClinVar VCF Normalize for instructions on how to download the raw ClinVar VCF and perform curation yourself.

Curated ClinVar TSV files are also available for download:

• ClinVar_PLP
• ClinVar_BLB

Alternatively, see ClinVar DataSet Curation for more details.

---

Annotators Installing

See Annotator Installing for installation instructions.

---

Variant Annotation

ANNOVAR

Prepare ANNOVAR avinput files:

cd /opt/annovar

INPUT_DIR=/your/clinvar/vcf/path/ClinVar_VCF
OUTPUT_DIR=/your/output/volume/ClinVar20240107/refseq_annovar

perl convert2annovar.pl --includeinfo -format vcf4 $INPUT_DIR/ClinVar_GRCh38_PLP_Tx1_20240107.vcf > $OUTPUT_DIR/GRCh38_PLP/ClinVar_GRCh38_PLP.avinput

perl convert2annovar.pl --includeinfo -format vcf4 $INPUT_DIR/ClinVar_GRCh38_BLB_Tx1_20240107.vcf > $OUTPUT_DIR/GRCh38_BLB/ClinVar_GRCh38_BLB.avinput

Run ANNOVAR with the RefSeq database (you can adjust the number of threads as needed):

IN_avinput=/your/output/volume/ClinVar20240107/refseq_annovar/GRCh38_PLP/ClinVar_GRCh38_PLP.avinput
OUT_DIR=/your/output/volume/ClinVar20240107/refseq_annovar/GRCh38_PLP

perl /volume/wgsa/annovar/table_annovar.pl $IN_avinput \
/volume/wgsa/annovar/humandb/ --buildver hg38 --protocol refGene --operation g -polish \
-out $OUT_DIR/ClinVar_GRCh38_PLP \
--thread 1 \
--argument '--neargene 5000 --hgvs --transcript_function --separate' \
--dot2underline --nastring .

IN_avinput=/your/output/volume/ClinVar20240107/refseq_annovar/GRCh38_BLB/ClinVar_GRCh38_BLB.avinput
OUT_DIR=/your/output/volume/ClinVar20240107/refseq_annovar/GRCh38_BLB

perl /volume/wgsa/annovar/table_annovar.pl $IN_avinput \
/volume/wgsa/annovar/humandb/ --buildver hg38 --protocol refGene --operation g -polish \
-out $OUT_DIR/ClinVar_GRCh38_BLB \
--thread 1 \
--argument '--neargene 5000 --hgvs --transcript_function --separate' \
--dot2underline --nastring .

SnpEff

Run SnpEff with the RefSeq database (you can adjust the Xmx RAM value as needed):

cd /opt/snpEff

INPUT_DIR=/your/clinvar/vcf/path/ClinVar_VCF
OUTPUT_DIR=/your/output/volume/ClinVar20240107/refseq_snpeff

SnpEff_INPUT_VCF=$INPUT_DIR/ClinVar_GRCh38_PLP_Tx1_20240107.vcf
SnpEff_OUTPUT_VCF=$OUTPUT_DIR/GRCh38_PLP/ClinVar_GRCh38_PLP_Tx1_20240107.snpeff.vcf

java -Xmx10g -jar snpEff.jar GRCh38.mane.1.2.refseq $SnpEff_INPUT_VCF > $SnpEff_OUTPUT_VCF

cd /opt/snpEff

INPUT_DIR=/your/clinvar/vcf/path/ClinVar_VCF
OUTPUT_DIR=/your/output/volume/ClinVar20240107/refseq_snpeff

SnpEff_INPUT_VCF=$INPUT_DIR/ClinVar_GRCh38_BLB_Tx1_20240107.vcf
SnpEff_OUTPUT_VCF=$OUTPUT_DIR/GRCh38_BLB/ClinVar_GRCh38_BLB_Tx1_20240107.snpeff.vcf

java -Xmx10g -jar snpEff.jar GRCh38.mane.1.2.refseq $SnpEff_INPUT_VCF > $SnpEff_OUTPUT_VCF

VEP

Run VEP with the RefSeq database (BUFFER_SIZE and FORK can be adjusted as needed):

cd /opt/ensembl-vep

INPUT_DIR=/your/clinvar/vcf/path/ClinVar_VCF
OUTPUT_DIR=//your/output/volume/ClinVar20240107/refseq_vep

VEP_INPUT_VCF=$INPUT_DIR/ClinVar_GRCh38_PLP_Tx1_20240107.vcf
VEP_OUTPUT_VCF=$OUTPUT_DIR/GRCh38_PLP/ClinVar_GRCh38_PLP_Tx1_20240107.vep.vcf

ASSEMBLY=GRCh38
BUFFER_SIZE=500
CACHE_VERSION=111
FORK=2

./vep \
  --format vcf \
  -i $VEP_INPUT_VCF \
  --vcf \
  -o $VEP_OUTPUT_VCF \
  --assembly ${ASSEMBLY} \
  --refseq \
  --offline \
  --cache \
  --cache_version ${CACHE_VERSION} \
  --dir_cache /opt/VepCache \
  --fasta /opt/VepCache/Homo_sapiens.GRCh38.dna.primary_assembly.fa.gz \
  --hgvsg \
  --hgvs \
  --total_length \
  --variant_class \
  --symbol \
  --protein \
  --canonical \
  --mane \
  --biotype \
  --no_stats \
  --no_escape \
  --buffer_size ${BUFFER_SIZE} \
  --fork ${FORK}

cd /opt/ensembl-vep

INPUT_DIR=/your/clinvar/vcf/path/ClinVar_VCF
OUTPUT_DIR=//your/output/volume/ClinVar20240107/refseq_vep

VEP_INPUT_VCF=$INPUT_DIR/ClinVar_GRCh38_BLB_Tx1_20240107.vcf
VEP_OUTPUT_VCF=$OUTPUT_DIR/GRCh38_PLP/ClinVar_GRCh38_BLB_Tx1_20240107.vep.vcf

ASSEMBLY=GRCh38
BUFFER_SIZE=500
CACHE_VERSION=111
FORK=2

./vep \
  --format vcf \
  -i $VEP_INPUT_VCF \
  --vcf \
  -o $VEP_OUTPUT_VCF \
  --assembly ${ASSEMBLY} \
  --refseq \
  --offline \
  --cache \
  --cache_version ${CACHE_VERSION} \
  --dir_cache /opt/VepCache \
  --fasta /opt/VepCache/Homo_sapiens.GRCh38.dna.primary_assembly.fa.gz \
  --hgvsg \
  --hgvs \
  --total_length \
  --variant_class \
  --symbol \
  --protein \
  --canonical \
  --mane \
  --biotype \
  --no_stats \
  --no_escape \
  --buffer_size ${BUFFER_SIZE} \
  --fork ${FORK}

---

Trrascript Curation After Annotation

We provide an R script to process and split transcripts following annotation by annotators. You can then use the supplied key together with the processed RefSeq transcript (NM) identifiers as join criteria for comparative analyses in the programming language of your choice.

• ANNOVAR Transcript Curation
• SnpEff Transcript Curation
• VEP Transcript Curation