Table 2 Independent TWAS risk genes of AD dementia identified by Avg-valid + SR models of SMA tissue

From: SR-TWAS: leveraging multiple reference panels to improve transcriptome-wide association study power by ensemble machine learning

  

Avg-valid + SR

SR-TWAS

PrediXcan

TIGAR

  

ROSMAP

SMA

ROSMAP

SMA

ROSMAP

SMA

ROSMAP

SMA

Gene

CHR

Z-score

P value

Z-score

P value

Z-score

P value

Z-score

P value

HLA-DRAac

6

−6.81

1.00e-11

−6.88

5.93e-12

-

-

-

-

AC092849.1bd

7

6.54

6.23e-11

-

-

6.48

9.02e-11

5.98

2.21e-09

SRD5A3P1bd

11

8.73

2.49e-18

-

-

-

-

8.73

2.49e-18

RN7SL225Pbd

11

5.05

4.33e-07

-

-

5.03

5.02e-07

4.93

8.04e-07

AL110118.2bd

14

−5.32

1.01e-07

-

-

−5.25

1.52e-07

−5.63

1.77e-08

FOSBbd

19

29.19

2.44e-187

-

-

29.20

1.87e-187

13.79

2.98e-43

  1. Two-sided TWAS Z-scores and p values were presented here for the models of Avg-valid + SR, SR-TWAS, PrediXcan validation base models, and TIGAR validation base models. The signs of Z-scores show the directions of the mediated genetic effects on AD dementia. The genome-wide significance threshold was Bonferroni-corrected for testing ~20 K genes.
  2. aKnown GWAS risk gene of AD.
  3. bGene within 1 Mb of known GWAS risk gene of AD.
  4. cPreviously observed TWAS risk gene of AD.
  5. dGene within 1 Mb of previously observed TWAS risk gene of AD.
Back to article page