.\" Automatically generated by Pandoc 2.7.3
.\"
.TH "IHS" "1" "" "iHS (vcflib)" "iHS (VCF statistics)"
.hy
.SH NAME
.PP
\f[B]iHS\f[R]
.SH SYNOPSIS
.PP
\f[B]iHS\f[R] \[en]target 0,1,2,3,4,5,6,7 \[en]file my.phased.vcf
\ \[en]region chr1:1-1000 > STDOUT 2> STDERR
.SH DESCRIPTION
.PP
\f[B]iHS\f[R] calculates the integrated haplotype score which measures
the relative decay of extended haplotype homozygosity (EHH) for the
reference and alternative alleles at a site (see: voight et al.\ 2006,
Spiech & Hernandez 2014).
.SH OPTIONS
.IP
.nf
\f[C]


Our code is highly concordant with both implementations mentioned. However, we do not set an upper limit to the allele frequency.  **iHS** can be run without a genetic map, in which case the change in EHH is integrated over a constant.  Human genetic maps for GRCh36 and GRCh37 (hg18 & hg19) can be found at: http://bochet.gcc.biostat.washington.edu/beagle/genetic_maps/ . **iHS** by default interpolates SNV positions to genetic position (you don\[aq]t need a genetic position for every VCF entry in the map file).

**iHS** analyses requires normalization by allele frequency.  It is important that **iHS** is calculated over large regions so that the normalization does not down weight real signals.  For genome-wide runs it is recommended to run slightly overlapping windows and throwing out values that fail integration (columns 7 & 8 in the output) and then removing duplicates by using the \[aq]sort\[aq] and \[aq]uniq\[aq] linux commands.  Normalization of the output is as simple as running \[aq]normalize-**iHS**\[aq].



     **iHS** calculates the integrated ratio of haplotype decay between the reference and non-reference allele.
Output : 4 columns :
     1. seqid
     2. position
     3. target allele frequency
     4. integrated EHH (alternative)
     5. integrated EHH (reference)
     6. **iHS** ln(iEHHalt/iEHHref)
     7. != 0 integration failure
     8. != 0 integration failure

Params:
       required: t,target  <STRING>  A zero base comma separated list of target
                                     individuals corresponding to VCF columns
       required: r,region  <STRING>  A tabix compliant genomic range
                                     format: \[dq]seqid:start-end\[dq] or \[dq]seqid\[dq]
       required: f,file    <STRING>  Proper formatted and phased VCF.
       required: y,type    <STRING>  Genotype likelihood format: GT,PL,GL,GP
       optional: a,af      <DOUBLE>  Alternative alleles with frquences less
                                     than [0.05] are skipped.
       optional: x,threads <INT>     Number of CPUS [1].
       recommended: g,gen <STRING>   A PLINK formatted map file.



Type: statistics


\f[R]
.fi
.SH EXIT VALUES
.TP
.B \f[B]0\f[R]
Success
.TP
.B \f[B]not 0\f[R]
Failure
.SH SEE ALSO
.PP
\f[B]vcflib\f[R](1)
.SH OTHER
.SH LICENSE
.PP
Copyright 2011-2020 (C) Erik Garrison and vcflib contributors.
MIT licensed.
.SH AUTHORS
Erik Garrison and vcflib contributors.