1 Association

analyses were conducted using 1,834 individulas with AF and 7

2 Microsatellite mutation rate

analyses were conducted using a drug-resistance reversion ass

3 Behavioral

analyses were conducted using a newly-developed force-sensing

4 Analyses were conducted using a random effect model.

5 Primary

analyses were conducted using all available data.

6 Six independent

analyses were conducted using all marker data (173 markers) a

7 Conditional logistic regression

analyses were conducted using clinically defined cutpoints, w

8 Bivariate and multivariable

analyses were conducted using conditional logistic regression

9 Single-locus

analyses were conducted using conditional logistic regression

10 hotic symptoms were determined by standardized methods, and

analyses were conducted using Cox regression modeling.

11 Analyses were conducted using data from 12 963 children aged

12 Analyses were conducted using data from 5998 adults in the Th

13 Linkage

analyses were conducted using data from a genome-wide 10-cM m

14 Data

analyses were conducted using generalized estimating equation

15 Patient-level

analyses were conducted using generalized estimating equation

16 Longitudinal

analyses were conducted using generalized estimating equation

17 Molecular pathway

analyses were conducted using Ingenuity Pathway Analysis tool

18 Survival

analyses were conducted using Kaplan-Meier estimates and mult

19 Analyses were conducted using linear regression to examine di

20 Statistical

analyses were conducted using linear regression.

21 The primary outcome

analyses were conducted using logistic regression analysis.

22 SNP-level

analyses were conducted using logistic regression models.

23 Heritability

analyses were conducted using maximum likelihood structural e

24 Linkage

analyses were conducted using model-free methods (SIBPAL, S.A

25 Analyses were conducted using multivariate repeated-measures

26 Single-nucleotide polymorphism (SNP)

analyses were conducted using PLINK with permutation procedur

27 The

analyses were conducted using QuEChERS extraction and gas chr

28 Meta-

analyses were conducted using random-effects modeling, and he

29 Meta-

analyses were conducted using random-effects models.

30 To extend these observations, complementation

analyses were conducted using S. flexneri SF621 and S. typhim

31 Genotypic resistance

analyses were conducted using Sanger population nucleotide se

32 riptive, bivariate, multivariate and Cochran-Armitage trend

analyses were conducted using SAS 9.2.

33 Analyses were conducted using SAS-callable SUDAAN to correct

34 Correlation

analyses were conducted using simple linear regression models

35 Global gene expression

analyses were conducted using splicing-sensitive microarrays

36 ive, Mixed Modelling and Generalised Linear Mixed Modelling

analyses were conducted using SPSS version 22.

37 All

analyses were conducted using standard microbiological and ch

38 Analyses were conducted using structural equation modeling fo

39 Mendelian randomization

analyses were conducted using summary statistics obtained for

40 Similar

analyses were conducted using Surveillance, Epidemiology, and

41 In addition, behavioral correlation

analyses were conducted using the Autism Diagnostic Interview

42 assess the consistency of apoE-CHD associations, stratified

analyses were conducted using the classic random-effects mode

43 estionnaires completed in 1992, 1994, 1996, 1998, and 2000;

analyses were conducted using the cumulative average of PA (m

44 Two-point and multipoint nonparametric linkage (NPL)

analyses were conducted using the NPL statistic and 1-paramet

45 Multiple meta-

analyses were conducted using the random effect model on raw

46 ualified, high-throughput, multiplexed immunoassays and all

analyses were conducted using the same software.

47 Quantitative genetic

analyses were conducted using the SOLAR analytic platform, a

48 Multivariable

analyses were conducted using the well validated GRACE (Globa

49 Whole-brain volumetric

analyses were conducted using voxel-based morphometry to dete

50 All

analyses were conducted using weighted data.