1. PCSK9 genetic variants and risk of type 2 diabetes: a mendelian randomisation study 
Dr Amand F Schmidt, PhD, Daniel I Swerdlow, PhD, Michael V Holmes, PhD, Riyaz S Patel, MD, Zammy Fairhurst-Hunter, MSc, Donald M Lyall, PhD, Fernando Pires Hartwig, MSc, Prof Bernardo Lessa Horta, PhD, Prof Elina Hyppönen, PhD, Prof Christine Power, PhD, Max Moldovan, PhD, Erik van Iperen, MSc, Prof G Kees Hovingh, PhD, Ilja Demuth, PhD, Kristina Norman, PhD, Prof Elisabeth Steinhagen-Thiessen, MD, Juri Demuth, PhD, Prof Lars Bertram, MD, Tian Liu, PhD, Stefan Coassin, PhD, Prof Johann Willeit, PhD, Stefan Kiechl, MD, Karin Willeit, MD, Dan Mason, PhD, Prof John Wright, FRCP, Prof Richard Morris, PhD, Prof Goya Wanamethee, PhD, Prof Peter Whincup, FRCP, Prof Yoav Ben-Shlomo, PhD, Stela McLachlan, PhD, Prof Jackie F Price, MD, Prof Mika Kivimaki, PhD, Catherine Welch, PhD, Adelaida Sanchez-Galvez, PhD, Pedro Marques-Vidal, PhD, Andrew Nicolaides, PhD, Andrie G Panayiotou, PhD, N Charlotte Onland-Moret, PhD, Prof Yvonne T van der Schouw, PhD, Giuseppe Matullo, PhD, Giovanni Fiorito, PhD, Simonetta Guarrera, MSc, Carlotta Sacerdote, PhD, Prof Nicholas J Wareham, PhD, Claudia Langenberg, PhD, Robert Scott, PhD, Jian'an Luan, PhD, Prof Martin Bobak, PhD, Prof Sofia Malyutina, PhD, Andrzej Pająk, PhD, Ruzena Kubinova, PhD, Prof Abdonas Tamosiunas, PhD, Hynek Pikhart, PhD, Lise Lotte Nystrup Husemoen, PhD, Niels Grarup, PhD, Oluf Pedersen, PhD, Torben Hansen, PhD, Prof Allan Linneberg, PhD, Kenneth Starup Simonsen, PhD, Jackie Cooper, MSc, Prof Steve E Humphries, PhD, Murray Brilliant, PhD, Terrie Kitchner, CCRP, Hakon Hakonarson, PhD, David S Carrell, PhD, Catherine A McCarty, PhD, H Lester Kirchner, PhD, Eric B Larson, MD, David R Crosslin, PhD, Prof Mariza de Andrade, PhD, Prof Dan M Roden, MD, Joshua C Denny, MD, Cara Carty, PhD, Stephen Hancock, MSciStud, John Attia, PhD, Elizabeth Holliday, PhD, Martin O'Donnell, PhD, Prof Salim Yusuf, DPhil, Michael Chong, MSc, Prof Guillaume Pare, MD, Prof Pim van der Harst, PhD, M Abdullah Said, BSc, Ruben N Eppinga, PhD, Niek Verweij, PhD, Prof Harold Snieder, PhD, Tim Christen, MSc, Dennis O Mook-Kanamori, PhD, Stefan Gustafsson, PhD, Prof Lars Lind, PhD, Prof Erik Ingelsson, PhD, Raha Pazoki, PhD, Oscar Franco, PhD, Prof Albert Hofman, PhD, Andre Uitterlinden, PhD, Abbas Dehghan, PhD, Alexander Teumer, PhD, Sebastian Baumeister, PhD, Prof Marcus Dörr, MD, Prof Markus M Lerch, MD, Prof Uwe Völker, PhD, Prof Henry Völzke, MD, Joey Ward, PhD, Jill P Pell, PhD, Daniel J Smith, PhD, Tom Meade, FRS, Prof Anke H Maitland-van der Zee, PhD, Ekaterina V Baranova, MSc, Robin Young, PhD, Ian Ford, PhD, Archie Campbell, MA, Prof Sandosh Padmanabhan, PhD, Prof Michiel L Bots, MD, Prof Diederick E Grobbee, PhD, Prof Philippe Froguel, PhD, Dorothée Thuillier, MSc, Beverley Balkau, PhD, Amélie Bonnefond, PhD, Prof Bertrand Cariou, MD, Melissa Smart, PhD, Yanchun Bao, PhD, Prof Meena Kumari, PhD, Anubha Mahajan, PhD, Prof Paul M Ridker, MD, Daniel I Chasman, PhD, Alex P Reiner, MD, Prof Leslie A Lange, PhD, Marylyn D Ritchie, PhD, Prof Folkert W Asselbergs, MD, Prof Juan-Pablo Casas, PhD, Brendan J Keating, PhD, David Preiss, MD, Prof Aroon D Hingorani, PhD, Prof Naveed Sattar, FMedSci 
The Lancet Diabetes & Endocrinology 
Volume 5, Issue 2, Pages (February 2017)
DOI: /S (16)
Copyright © 2017 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY license Terms and Conditions

2. Figure 1
Association of genetic variants in PCSK9 with circulating LDL cholesterol concentration
Effect estimates are presented as mean difference in LDL cholesterol (mmol/L) per LDL cholesterol-lowering allele, with 95% CIs. Results are pooled by use of a fixed-effect model. The size of the black dots representing the point estimates is proportional to the inverse of the variance. Note that results from individual participant data are supplemented by repository data from the Global Lipids Genetics Consortium.
The Lancet Diabetes & Endocrinology 2017 5, DOI: ( /S (16) )
Copyright © 2017 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY license Terms and Conditions

3. Figure 2
Association of genetic variants in PCSK9 with glycaemic and anthropometric biomarkers
Effect estimates are presented as mean difference with 95% CIs. Associations were scaled to a 1 mmol/L reduction in LDL cholesterol. SNP-specific results are pooled by use of a fixed-effect model; weighted gene-centric score (GS) models combining all four SNP-specific estimates are presented as fixed-effect and random-effects estimates. The size of the black dots representing the point estimates is proportional to the inverse of the variance. Between-SNP heterogeneity was measured as a two-sided Q-test (χ2) and an I2 with one-sided 97·5% CI. Note that results from individual participant data are supplemented by repository data from the Global Lipids Genetics Consortium, the Meta-Analyses of Glucose and Insulin-related traits Consortium, and the Genetic Investigation of Anthropometric Traits consortium.
The Lancet Diabetes & Endocrinology 2017 5, DOI: ( /S (16) )
Copyright © 2017 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY license Terms and Conditions

4. Figure 3
Association of genetic variants in PCSK9 with risk of type 2 diabetes, individually (A) and as weighted gene-centric score (B)
Effect estimates are presented as odds ratios (ORs) for the incidence or prevalence of type 2 diabetes, with 95% CIs. Associations were scaled to a 1 mmol/L reduction in LDL cholesterol. SNP-specific results are pooled by use of a fixed-effect model; weighted gene-centric score (GS) models combining all four SNP-specific estimates are presented as fixed-effect and random-effects estimates. The size of the black dots representing the point estimates is proportional to the inverse of the variance. Between-SNP heterogeneity was measured as a two-sided Q-test (χ2) and an I2 with one-sided 97·5% CI. Results from individual participant data are supplemented by repository data from the Diabetes Genetics Replication and Meta-analysis consortium.
The Lancet Diabetes & Endocrinology 2017 5, DOI: ( /S (16) )
Copyright © 2017 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY license Terms and Conditions

5. Figure 4
Correlation between PCSK9 associations with LDL cholesterol concentration and type 2 diabetes
Effect estimates are presented as mean difference in LDL cholesterol concentration (mmol/L) and odds ratios (ORs) for the incidence or prevalence of type 2 diabetes, with 95% CIs. Associations are presented per LDL cholesterol-decreasing allele. The Pearson correlation coefficient, regression line (grey), and its 95% CI (red) were calculated by weighting the SNPs for the inverse of the variance in the type 2 diabetes association. Excluding the SNP with the largest effect on LDL cholesterol (rs ) resulted in a correlation coefficient of 0·993 and a p value of 0·437.
The Lancet Diabetes & Endocrinology 2017 5, DOI: ( /S (16) )
Copyright © 2017 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY license Terms and Conditions