Motivation: Microarray data must be background corrected to remove the effects of non-specific binding or spatial heterogeneity across the array, but this practice typically causes other problems such as negative corrected intensities and high variability of low intensity log-ratios. Different estimators of background, and various model-based processing methods, are compared in this study in search of the best option for differential expression analyses of small microarray experiments. Results: Using data where some independent truth in gene expression is known, eight different background correction alternatives are compared, in terms of precision and bias of the resulting gene expression measures, and in terms of their ability to detect differentially expressed genes as judged by two popular algorithms, SAM and limma eBayes. A new background processing method (normexp) is introduced which is based on a convolution model. The model-based correction methods are shown to be markedly superior to the usual practice of subtracting local background estimates. Methods which stabilize the variances of the log-ratios along the intensity range perform the best. The normexp+offset method is found to give the lowest false discovery rate overall, followed by morph and vsn. Like vsn, normexp is applicable to most types of two-colour microarray data. Availability: The background correction methods compared in this article are available in the R package limma (Smyth, 2005) from http://www.bioconductor.org. Contact:smyth@wehi.edu.au Supplementary information: Supplementary data are available from http://bioinf.wehi.edu.au/resources/webReferences.html.

A multidisciplinary collaboration investigated the world's largest, most catastrophic epidemic thunderstorm asthma event that took place in Melbourne, Australia, on Nov 21, 2016, to inform mechanisms and preventive strategies. Meteorological and airborne pollen data, satellite-derived vegetation index, ambulance callouts, emergency department presentations, and data on hospital admissions for Nov 21, 2016, as well as leading up to and following the event were collected between Nov 21, 2016, and March 31, 2017, and analysed. We contacted patients who presented during the epidemic thunderstorm asthma event at eight metropolitan health services (each including up to three hospitals) via telephone questionnaire to determine patient characteristics, and investigated outcomes of intensive care unit (ICU) admissions. Grass pollen concentrations on Nov 21, 2016, were extremely high (>100 grains/m3). At 1800 AEDT, a gust front crossed Melbourne, plunging temperatures 10°C, raising humidity above 70%, and concentrating particulate matter. Within 30 h, there were 3365 (672%) excess respiratory-related presentations to emergency departments, and 476 (992%) excess asthma-related admissions to hospital, especially individuals of Indian or Sri Lankan birth (10% vs 1%, p<0·0001) and south-east Asian birth (8% vs 1%, p<0·0001) compared with previous 3 years. Questionnaire data from 1435 (64%) of 2248 emergency department presentations showed a mean age of 32·0 years (SD 18·6), 56% of whom were male. Only 28% had current doctor-diagnosed asthma. 39% of the presentations were of Asian or Indian ethnicity (25% of the Melbourne population were of this ethnicity according to the 2016 census, relative risk [RR] 1·93, 95% CI 1·74-2·15, p <0·0001). Of ten individuals who died, six were Asian or Indian (RR 4·54, 95% CI 1·28-16·09; p=0·01). 35 individuals were admitted to an intensive care unit, all had asthma, 12 took inhaled preventers, and five died. Convergent environmental factors triggered a thunderstorm asthma epidemic of unprecedented magnitude, tempo, and geographical range and severity on Nov 21, 2016, creating a new benchmark for emergency and health service escalation. Asian or Indian ethnicity and current doctor-diagnosed asthma portended life-threatening exacerbations such as those requiring admission to an ICU. Overall, the findings provide important public health lessons applicable to future event forecasting, health care response coordination, protection of at-risk populations, and medical management of epidemic thunderstorm asthma. None.