Plots of <i>P</i>-values to evaluate many tests simultaneously

Schweder, Tore; Spjøtvoll, Emil

doi:10.1093/biomet/69.3.493

Cited by 374 publications

(201 citation statements)

References 14 publications

Supporting

Mentioning

199

Contrasting

Order By: Relevance

“…However, in certain circumstances, the Bonferroni procedure may be too conservative. We have therefore followed Schweder and Spjotvoll (1982) and plotted the cumulative number of P-values, Cp, against (1-P). The slope of the straight line drawn in Figure 4 is approximately 384, which, if taken as the number of 'true' null hypotheses, suggests that there could be 17 districts in which differences in SYPLR might not have arisen entirely by chance.…”

Section: Resultsmentioning

confidence: 99%

“…Only two districts, Bradford and Leeds, have a common boundary. Schweder and Spjotvoll (1982) say that their P-value plot '... is primarily intended for informal inference and it is difficult to make exact probability statements.' As an approximate test of significance they recommend, in cases such as this, using ass = 0.05/384 = 0.00013, which is only slightly higher than the conventional Bonferroni figure and would still result in no districts with statistically significant SYPLR.…”

Section: Resultsmentioning

confidence: 99%

“…Use of the Bonferroni correction, even when modified according to Schweder and Spjotvoll (1982), would suggest that practically all could be attributed to sampling variation. This is probably a very conservative conclusion as the plot in Figure 4 indicates that there are approximately 17 null hypotheses which should be rejected.…”

Section: Discussionmentioning

confidence: 99%

“…An alternative approach first calculates the P-values for the n comparisons involved (Schweder and Spjotvoll, 1982;Haybittle et al, 1995). These values are ranked and then plotted against the cumulative number of tests conducted.…”

Section: Multiple Comparisonsmentioning

confidence: 99%

See 3 more Smart Citations

Geographical variation in the standardized years of potential life lost ratio (SYPLR) in women dying from malignancies of the breast in England and Wales

Yuen¹,

Haybittle²,

Machin³

1997

Br J Cancer

View full text Add to dashboard Cite

Summary Geographical variation in the standardized years of potential life lost ratio (SYPLR) in women aged dying from breast cancer has been mapped by local authority district in England and Wales and compared with the variation as described by the standardized mortality ratio (SMR). The geographical distribution of areas of low and high SMRs is similar to that observed some 15 years earlier, showing an increase from north to south. In contrast, the pattern of SYPLRs shows a less obvious trend. Years of life lost because of breast cancer mortality were particularly high in eight of the 401 county districts (SYPLR 1.22-1.48) and particularly low in nine (0.60-0.86 In England, the prevention and early detection of breast cancer in women is one of the five key areas for the Health of the Nation, an initiative put forward by the British government with the strategic goal of improving health through prevention and promotion (Department of Health, 1992). In England and Wales, 1 in 12 women develops breast cancer and, in 1988 alone, almost 27 000 new cases were diagnosed. This is approximately double the number of new cases of the next most common tumours, i.e. colorectal and lung, which affect both sexes (Thames Cancer Registry, 1994). Of the 13 663 women who died from breast cancer in 1992, 60% were under 75 years of age (Office of Population Censuses and Surveys, 1994).In setting up objectives and target outcome for cancer as one of the key areas for Health of the Nation action, mortality and cancer incidence data were used to obtain age-standardized indicators However, there has been some concern over the use of the standardized mortality ratio (SMR) in this context.Thus, the use of years of potential life lost (YPL) has been prop.osed (Dempsey, 1947; Greville, 1948;Haesnzel, 1950). This measure provides an indicator which gives more weight to deaths occurring at younger ages, thus emphasizing the impact of premature mortality (Romeder and McWhinnie, 1977). Such a measure reflects, in some sense, consequent social, family and economic burdens. It has been employed to investigate variation in accidents (Centres for Disease Control, 1991) and AIDS/HIV (Saunders et al, 1990) as well as cancer (Stocks, 1953;Horm and Sondik, 1989 Mapping of disease indicators has always been an important tool for epidemiologists, public health physicians and health care planners, especially when examining the geographic variation in cancers (Howe, 1989;Stocks, 1936) for clues to aetiology or to facilitate allocation of resources.In this paper, we use the standardized years of potential life lost ratio (Haesnzel, 1950) (SYPLR), denoted the SYLL ratio by Marlow (1995), as a summary measure of the impact of premature mortality from breast cancer in women aged 20-74 years and investigate its geographical variation in England and Wales among 403 local authority districts. This age group was chosen as death from breast cancer in women under 20 is rare and the majority of women remain independent (healthwise) up to the age of 75. We also ...

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Multiple Comparisonsmentioning

confidence: 99%

See 2 more Smart Citations

Geographical variation in the standardized years of potential life lost ratio (SYPLR) in women dying from malignancies of the breast in England and Wales

Yuen¹,

Haybittle²,

Machin³

1997

Br J Cancer

View full text Add to dashboard Cite

show abstract

“…A priori we do not know how many true associations to expect, even though it is reasonable to assume that many co-occurrences are indeed meaningful, given the manual curation process of the MeSH metadata. In order to rationalize the choice of a significance threshold, we use a method specifically developed for a similar application 61 that combines w 2 -tests with P-value plots 62 (see Supplementary Methods and Supplementary Fig. 3 for more details).…”

Section: Methodsmentioning

confidence: 99%

Human symptoms–disease network

Zhou¹,

Menche²,

Barabási³

et al. 2014

Nat Commun

600

484

View full text Add to dashboard Cite

In the post-genomic era, the elucidation of the relationship between the molecular origins of diseases and their resulting phenotypes is a crucial task for medical research. Here, we use a large-scale biomedical literature database to construct a symptom-based human disease network and investigate the connection between clinical manifestations of diseases and their underlying molecular interactions. We find that the symptom-based similarity of two diseases correlates strongly with the number of shared genetic associations and the extent to which their associated proteins interact. Moreover, the diversity of the clinical manifestations of a disease can be related to the connectivity patterns of the underlying protein interaction network. The comprehensive, high-quality map of disease-symptom relations can further be used as a resource helping to address important questions in the field of systems medicine, for example, the identification of unexpected associations between diseases, disease etiology research or drug design.

show abstract

Genome-wide Pleiotropy Between Parkinson Disease and Autoimmune Diseases

et al. 2017

View full text Add to dashboard Cite

IMPORTANCE Recent genome-wide association studies (GWAS) and pathway analyses supported long-standing observations of an association between immune-mediated diseases and Parkinson disease (PD). The post-GWAS era provides an opportunity for cross-phenotype analyses between different complex phenotypes. OBJECTIVES To test the hypothesis that there are common genetic risk variants conveying risk of both PD and autoimmune diseases (ie, pleiotropy) and to identify new shared genetic variants and their pathways by applying a novel statistical framework in a genome-wide approach. DESIGN, SETTING, AND PARTICIPANTS Using the conjunction false discovery rate method, this study analyzed GWAS data from a selection of archetypal autoimmune diseases among 138 511 individuals of European ancestry and systemically investigated pleiotropy between PD and type 1 diabetes, Crohn disease, ulcerative colitis, rheumatoid arthritis, celiac disease, psoriasis, and multiple sclerosis. NeuroX data (6927 PD cases and 6108 controls) were used for replication. The study investigated the biological correlation between the top loci through protein-protein interaction and changes in the gene expression and methylation levels. The dates of the analysis were June 10, 2015, to March 4, 2017. MAIN OUTCOMES AND MEASURES The primary outcome was a list of novel loci and their pathways involved in PD and autoimmune diseases. RESULTS Genome-wide conjunctional analysis identified 17 novel loci at false discovery rate less than 0.05 with overlap between PD and autoimmune diseases, including known PD loci adjacent to GAK, HLA-DRB5, LRRK2, and MAPT for rheumatoid arthritis, ulcerative colitis and Crohn disease. Replication confirmed the involvement of HLA, LRRK2, MAPT, TRIM10, and SETD1A in PD. Among the novel genes discovered, WNT3, KANSL1, CRHR1, BOLA2, and GUCY1A3 are within a protein-protein interaction network with known PD genes. A subset of novel loci was significantly associated with changes in methylation or expression levels of adjacent genes. CONCLUSIONS AND RELEVANCE The study findings provide novel mechanistic insights into PD and autoimmune diseases and identify a common genetic pathway between these phenotypes. The results may have implications for future therapeutic trials involving anti-inflammatory agents.

show abstract

Plots of P-values to evaluate many tests simultaneously

Cited by 374 publications

References 14 publications

Geographical variation in the standardized years of potential life lost ratio (SYPLR) in women dying from malignancies of the breast in England and Wales

Geographical variation in the standardized years of potential life lost ratio (SYPLR) in women dying from malignancies of the breast in England and Wales

Human symptoms–disease network

Genome-wide Pleiotropy Between Parkinson Disease and Autoimmune Diseases

Contact Info

Product

Resources

About