Gene Scores - Unadjusted - Regular

Gene scores for selected combinations of phenotypes andSNV-to-gene mappings as calculated using genuine summary statistics with MAGMA's (v1.08) SNP-Wise Mean algorithm. <br>Refer to MAGMA's original publication as well as to theMAGMA website (https://ctg.cncr.nl/software/magma) for details concerning the SNP-Wise mean algorithm and the significance of MAGMA's v1.08 update in this regard.<br><b>NOTE:</b> We refer to the gene scores in this directory as "unadjusted", since they are not yet adjusted (following truncation of outlying gene scores) for residual effects of SNV count, within-gene LD, inverse mean minor-allele count, and sample size. Refer to the MAGMA user manual (downloadable from the MAGMA website) for details. In our analyses, unadjusted gene scores served merely as input to MAGMA's gene-set analysis in order to obtain (i) adjusted gene scores and (ii) gene-set scores (based on enrichment testing with adjusted gene scores). Refer to our publication for details. We supply these files forothers to execute gene-set analyses with precomputed genescores for diverse combinations of phenotypes and mappings. <br>File names are descriptive. For example:<br>GENEBODY_U2D2_EPM-GENEHANCER_DEFAULT-SUMSTATS_G1000EUR_MAF1<br>(1) GENEBODY - we mapped SNVs to genes with which they overlap (irrespective of whether the overlapwas with an intron or an exon).(2) U2D2 - flanks for mapping proximal SNVs to genes.U (flank upstream of the transcription start-site) and D (flank downstream of the transcription end-site) where the integers specify the corresponding flank's size in kb.(3) EPM-GENEHANCER - name of dataset of regulatoryinteractions used for augmentation of SNV-to-gene mapping, following the naming system used in our publication (NONEmeans no augmentation with RIs).(4) DEFAULT-SUMSTATS - specifies that these scoreswere obtained with genuine summary statistics andnot with any permutations (that is, not with theEPVP permutation control discussed in our publication).(5) G1000EUR - specifies that we used the 1000 Genomes Project European Reference Population files (that is,those downloadable from the MAGMA website) for mappingSNVs to genes (see next point as well).(6) MAF1 - specifically, we used SNVs with a minorallele frequency (MAF) of 1% from the relevantMAGMA binary files to build our mappings. Onecan use a software such as PLINK to filter thebinary files according to properties such as MAF(https://zzz.bwh.harvard.edu/plink/).<br>We provide both the "genes.out" and "genes.raw" files.These are equivalent in terms of gene scores but the"genes.raw" file format has a less-friendly format andserves as input to gene-set analysis (that is, this filecontains additional information on gene-gene correlationsbased on linkage disequilibrium).