Beagle(1)

🧬 Beagle(1) 💖

okay so bestie, welcome to the Beagle manual page — she's a General Commands Manual girlie and she does genotype calling, genotype phasing, and imputation of ungenotyped markers. pull up a chair, we're gonna get into it. ✨

---

🌸 NAME 🌸

Beagle — genotype calling, genotype phasing and imputation of ungenotyped markers. she's booked, busy, and iconic for it. 👑

---

✨ SYNOPSIS ✨

java -Xmx[GB]g -jar /usr/share/beagle/beagle.jar [options]

---

🦋 DESCRIPTION 🦋

so what does she actually do, you ask? Beagle performs genotype calling, genotype phasing, imputation of ungenotyped markers, and identity-by-descent segment detection. literally a quadruple threat. 💅

the genotypic imputation? it works on phased haplotypes using a Li and Stephens haplotype frequency model. no cap, that's the whole vibe.

and Beagle also implements the Refined IBD algorithm for detecting homozygosity-by-descent (HBD) and identity-by-descent (IBD) segments. she contains multitudes fr fr. 🌈

---

🎀 OPTIONS 🎀

💾 Data input/output parameters

these are the ones that tell Beagle where stuff comes from and where it goes. the plumbing, the logistics, the main character of file handling. 🫶

• gt=filename — Optional. Specifies a VCF file containing a GT (genotype) format field for each marker. if a genotype contains the phased allele separator, "|", then Beagle will preserve the phase of the genotype during the analysis. and heads up bestie: if you use the gt argument, all genotypes in the output file will be phased and non-missing. that's not a suggestion, that's just what happens. 😌
• gl=filename — Optional. Specifies a VCF file containing a GL or PL (genotype likelihood) format field for each marker. any data in the GT format field? ignored, sorry girl. and if both GL and PL format fields are present for a marker, the GL format will be used. GL stays winning. 💅
• gtgl=filename — Optional. Specifies a VCF file containing a GT, GL or PL format field for each marker. if a genotype is non-missing, Beagle will ignore the genotype likelihood. and if both GL and PL format fields are present for a marker, the GL field will be used. same energy as above, she's consistent. ✨
• ref=filename — Optional. Specifies a VCF file containing phased reference genotypes. go check the impute parameter if you wanna know more, iykyk. 👀
• out=prefix — Required. Specifies the output filename prefix. the prefix may be an absolute or relative filename, but it cannot be a directory name. do not try it. she will not be doing that. 🙈
• excludesamples=filename — Optional. Specifies a file containing non-reference samples (one sample per line) to be excluded from the analysis and output files. bye girl, you're not in this scene. 👋
• excludemarkers=filename — Optional. Specifies a file containing markers (one marker per line) to be excluded from the analysis and the output files. an excluded marker identifier can either be an identifier from the VCF record's ID field or a genomic coordinate in the format: CHROM:POS. two ways to get cut, pick your fighter. ✂️
• map=filename — Optional. Specifies a PLINK format genetic map on the cM scale. HapMap GrCh36 and GrCh37 genetic maps in PLINK format are available for download from the Beagle website. use of a genetic map is recommended if you are imputing ungenotyped markers. and if no genetic map is specified, Beagle will assume a constant recombination rate of 1 cM / Mb. default behavior, she's chill like that. 🌸
• chrom=chrom:start-end — Optional. Specifies a chromosome or chromosome interval using a chromosome identifier in the VCF file and the starting and ending positions of the interval. the entire chromosome, the beginning of the chromosome, and the end of a chromosome can be specified by chrom=[chrom], chrom=[chrom:-end], and chrom=[chrom:start-] respectively. pick your slice, bestie. 🍰
• maxlr=number_≥_1 — Default = 5000. Specifies the maximum likelihood ratio at a genotype. if M is the maximum of the likelihoods of each possible genotype, any likelihood that is less than (M ⁄ maxlr) is set to 0.0 to improve computational efficiency. she's just trimming the dead weight, as she should. 🔥

⚙️ General parameters

the settings that govern how she runs, how much RAM she's eating, and how many cores she's flexing. 💪

• nthreads=positive_integer — Default: machine-dependent. Specifies the number of threads of execution. if no nthreads parameter is specified, the nthreads parameter will be set equal to the number of CPU cores on the host machine. she'll use everything you've got, no permission needed. 😤
• lowmem=true/false — Default = false. Specifies whether a memory efficient algorithm should be used. the memory efficient algorithm increases run-time by a factor less than 2.0. so like, slower but lighter. a tradeoff, a little deal with the universe. 🧚
• window=positive_integer — Default = 50000. Specifies the number of markers to include in each sliding window. the window parameter must be at least twice as large as the overlap parameter. the window parameter controls the amount of memory used in the analysis. for human data, it is recommended that the window parameter be greater than or equal to the typical number of markers in 5 cM. math girlies stay winning. 📐
• overlap=positive_integer — Default = 3000. Specifies the number of markers of overlap between sliding windows. for human data, it is recommended that the overlap be set to the typical number of markers in 0.5 cM (when ibd=false) or 2.0 cM (when ibd=true). yes the setting changes based on another setting, she's layered. 💅
• seed=integer — Default = -99999. Specifies the seed for the random number generator. chaotic but reproducible. 🎲

🧬 Phasing and imputation parameters

the actual science-y ones. this is where the magic happens, and honestly? she ate. 🔥

• niterations=non-negative_integer — Default = 5. Specifies the number of phasing iterations. the phasing iterations are preceded by 10 burn-in iterations which carry out the Beagle version 4.0 phasing algorithm. if you want to phase your data with the Beagle 4.0 phasing algorithm, use niterations=0. accuracy and compute time increase with the number of iterations. more iterations, more time, more accuracy. the eternal triangle. 😭
• impute=true/false — Default = true. Specifies whether markers that are present in the reference panel but absent in your data will be imputed. this option has no effect if the ref and gt arguments are not used. read the fine print bestie. 📝
• gprobs=true/false — Default = false. Specifies whether a GP (genotype probability) format field will be included in the output VCF file when imputing ungenotyped markers. by default, a GP field is not printed because a DS (alternate allele dose) format field is always printed when imputing ungenotyped markers. DS is the default main character. 👑
• ne=integer — Default = 1000000. Specifies the effective population size when imputing ungenotyped markers. the default value is suitable for a large outbred human population. smaller values in the hundreds or thousands for the ne parameter are suggested for inbred human and animal populations. know your audience, adjust accordingly. 🐾
• err=non-negative_number — Default = 0.0001. Specifies the allele miscall rate. the default value should give good results for most sequence and SNP array data. she's pre-tuned, you're welcome. ✨
• cluster=non-negative_number — Default = 0.005. Specifies the maximum cM distance between individual markers that are combined into an aggregate marker when imputing ungenotyped markers. grouping them up, little marker friend group. 🫂

🔍 IBD parameters

the identity-by-descent section. relatedness detection, the family tree reveal moment. 🌳

• ibd=true/false — Default = false. Specifies whether IBD analysis will be performed when the gt argument is used. 📊
• ibdlod=non-negative_integer — Default = 3.0. Specifies the minimum LOD score for reported IBD. 🔬
• ibdscale=non-negative_number — Default: data-dependent. Specifies the scale parameter used to build the haplotype frequency model for IBD analysis. if no ibdscale parameter is specified, the scale parameter for the IBD analysis will be set to max{2, sqrt[sample size]/100}, which we have found to work well for outbred populations. she did the math so you don't have to. 🧮
• ibdtrim=non-negative_integer — Default = 40. Specifies the number of markers trimmed from the end of a shared haplotype when testing for IBD.

note: the default ibdtrim parameter is designed for European samples genotyped with a 1M SNP array (~ 1 marker per 3 kb). for human SNP array data, it is recommended to set the ibdtrim parameter to the typical number of markers in a 0.15 cM region. pilot studies of randomly selected genomic regions can be used to fine-tune the values of the ibdtrim parameter. so like... do your homework, but make it science. 📚

---

👀 SEE ALSO 👀

• https://faculty.washington.edu/browning/beagle/beagle.html

---

💜 AUTHOR 💜

Beagle was written by Brian L. Browning. 🫶

This manual page was written by Dylan Aïssi <bob.dybian@gmail.com>, for the Debian project (but may be used by others). she's for everyone, generous queen. 💖

---

okay that's the whole page bestie, you're officially educated. this was Beagle 4.1, dated February 2016, and honestly? iconic behavior from start to finish. go phase something. periodt. 💅✨🎀🦋🌸