LDscore Input Clarification

[borangao]

Dear Liyang,

Congratulations on gsMap—it is such a great tool. I have a clarification question regarding the use of LD scores. In the Mouse Embryo (Step-by-Step) tutorial, it seems that in Step 4 the LD scores are taken from the provided resource (gsMap_resource/LDSC_resource/weights_hm3_no_hla/weights.), rather than those computed from the 1000G data in Step 3. However, I do see the computed LD score files in (E16.5_E1S1.MOSTA/generate_ldscore/w_ld/weights.). I compared the two sets of LD scores and found them to be close, though not exactly the same.

Could you clarify which LD scores are intended to be used for the analysis?

Thank you very much for your time and help.

Best,
Boran

[LeonSong1995]

Hi Boran,

You can use either of them, as the results should be very similar.

This weight file was used to account for the correlation of chi² statistics between SNPs in LDSC. Strictly speaking, the weight file should match the SNP set used in the input GWAS data. For human GWAS data, we default to the HapMap3 (HM3) SNPs and the 1000 Genomes panel; therefore, the file LDSC_resource/weights_hm3_no_hla/weights should be used. For GWAS data from other species, where well-defined SNP sets are generally unavailable, we generated the weight file using SNPs with MAF > 0.05.

Best regards,
Liyang