Haplotype Analysis of Phased Genotype Data

Question

[alakatos]

Dear All,

I am a complete novice to haplotype analysis and I spent days to search for a tutorial or any guidance. No luck.

I have 400 subjects ( 200 CTRLs and 200 Cases) genotyped on an illumina platform. I used my data in PLINK files to phase my subjects with SHAPEIT (chr by chr).
The output format is hap and sample files.

How can be these files analyzed and visualized? How to move further?

Options based on my search:

1. It seems "HAPLOVIEW" can take only "HAPMAP PHASE" format: hap, legend and sample. I was wondering if there is a way to convert SHAPEIT output format to HAPMAP format? What tools can subset a HAPMAP format or SHAPEIT output? I need to analyze only a 200Kb region of chr11.

2. PLINK1.07 can handle haplotype analysis. However, phasing with PLINK is not recommended. Is there a way to integrate the phased files from SHAPEIT with PLINK.

Is there a way to create "myfile.hlist" from SHAPEIT output for PLINK.

PLINK document:

Haplotype testing in PLINK requires that the user supplies a file listing the haplotypes to be tested (Some precomputed lists are given below which might be useful in some circumstances.) The formats of these files are described below. An alternative is to specify a simple, sliding window of fixed haplotype size (also described below). The command: plink --file mydata --hap myfile.hlist

1. I can use PLINK --recodeHV option to make a linkage file for HAPLOVIEW. How can I phase it? The HAPMAP or GeneCruise tracks are not available anymore for download.

2. Can I convert the SHAPEIT phased hap and sample files to PLINK files? Would be these files phased PLINK files that could be used with the --block option and other association analysis without the need of "myfile.hlist"?

Sorry for the lenghty post but I would like to really understand haplotype analysis and do it in the proper way.

Any help, guidance, tutorial and recommendation of tools are highly appreciated. Thanks a lot, Anita

[Fabio Marroni]

For sure you can "easily" convert shapeit results to plink format. I don't see any way in which you can avoid the use of myfile.hlist. Also, I am not sure if there is a way to avoid that plink performs his E-M phasing, which you do not want.

I suspect that the reason why not much information on haplotype-based association analysis is available, is that the results are not so exciting.

Finally I suggest you read this paper by Browning (hte author of Beagle) on association analysis using phased data. Maybe Beagle itself can do that.

[alakatos]

Thank you for your help. Anita From: Fabio Marroni on Biostar [e-mail] Reply-To: "[e-mail]" <[e-mail]> Date: Friday, January 27, 2017 at 3:44 AM To: "Lakatos, Anita" [e-mail] Subject: [biostar] Haplotype Analysis of Phased Genotype Data

Activity on a post you are following on Biostarhttp://www.biostars.org

User Fabio MarroniFabio Marroni wrote Comment: Haplotype Analysis of Phased Genotype Data C: Haplotype Analysis of Phased Genotype Data :

For sure you can "easily" convert shapeit results to plink format. I don't see any way in which you can avoid the use of myfile.hlist. Also, I am not sure if there is a way to avoid that plink performs his E-M phasing, which you do not want.

Finally, I suspect that the reason why not much information on haplotype-based association analysis is available, is that the results are not so exciting.

Finally I suggest you read this paperhttps://www.ncbi.nlm.nih.gov/pmc/articles/PMC1474089/ by Browning (hte author of Beagle) on association analysis using phased data. Maybe Beagle itself can do that.

[Hernán]

"PLINK1.07" : Please note that unless very specific reasons, you should always use the most updated stable release from any software. Otherwise you wouldn't get support for questions or feature requests, etc.

There is PLINK 1.9 at: https://www.cog-genomics.org/plink2/

The command line to recode to HaploView from Plink is something like:

plink --recode hv --file ped_name --out hv_name --make-bed

[Hernán]

Nevermind, I just read from the homepage they obsoleted Haploype reading feature: "PLINK's haplotype phasing and imputation algorithms are obsolete. Future PLINK versions will be able to import phase and dosage information emitted by other programs; the haplotype association commands will be reintroduced when that functionality is in place. Until then, BEAGLE 3.3 should be more accurate than PLINK for case/control haplotype association."