Design and analysis of two-color microarray experiments using linear models.

OBJECTIVES: A variety of linear models have recently been proposed for the design and analysis of microarray experiments. This article gives an introduction to the most common models and describes their respective characteristics. METHODS: We focus on the application of linear models to logarithmized and normalized microarray data from two-color arrays. Linear models can be applied at different stages of evaluating microarray experiments, such as experimental design, background correction, normalization and hypothesis testing. Both one-stage and two-stage linear models including technical and possibly biological replicates are described. Issues related to selecting robust and efficient microarray designs are also discussed. RESULTS: Linear models provide flexible and powerful tools, which are easily implemented and interpreted. The methods are illustrated with an experiment performed in our laboratory, which demonstrates the value of using linear models for the evaluation of current microarray experiments. CONCLUSIONS: Linear models provide a flexible approach to properly account for variability, both across and within genes. This allows the experimenter to adequately model the sources of variability, which are assumed to be of major influence on the final measurements. In addition, design considerations essential for any well-planned microarray experiments are best incorporated using linear models. Results from such experimental design investigations show that the widely used common reference design is often substantially less efficient than alternative designs and its use is therefore not recommended.

Multiplicity issues in microarray experiments.

OBJECTIVES: Discussion of different error concepts relevant to microarray experiments. Review of some commonly used multiple testing procedures. Comparison of different approaches as applied to gene expression data. METHODS: This article focuses on familywise error rate (FWER) and false discovery rate (FDR) controlling procedures. Methods under investigation include: Bonferroni-type methods and their improvements (including resampling approaches), modified Bonferroni methods, data-driven approaches, as well as the linear step-up method and its modifications. Particular emphasis lies on the description of the assumptions, advantages and limitations for the investigated methods. RESULTS: FWER controlling procedures are often too conservative in high dimensional screening studies. A better balance between the raw P-values and the stringent FWER-adjusted P-values may be required in many situations, as provided by FDR controlling and related procedures. CONCLUSIONS: The questions remain open, which error concept to apply and which multiple testing procedure to use. Although we believe that the FDR or one of its variants will be applied more often in the future, longterm experience with microarray technology is missing and thus the validity of appropriate multiple test procedures cannot yet be assessed for microarray data analysis.

Regulation of the hypothalamic-pituitary-adrenocortical system in mice deficient for CRH receptors 1 and 2.

Recent investigations in mouse lines either deficient for the CRH receptor 1 (CRHR1) or 2 (CRHR2) suggest that the CRH neuronal system may comprise two separate pathways that can be coordinately and inversely activated in stress-induced hypothalamic-pituitary-adrenal (HPA) response and anxiety-like behavior. We generated mice deficient for both CRHR1 (Crhr1(-/-)) and CRHR2 (Crhr2(-/-)) to investigate the HPA system regulation in the absence of known functionally active CRH receptors under basal conditions and in response to different ethologically relevant stressors. To elucidate possible gene dose effects on the action of both CRH receptors, our analysis included heterozygous and homozygous CRHR1- or CRHR2-deficient mice, mutants lacking both CRH receptors, compound mutants with homozygous and heterozygous deficiency for either of the receptors, and their wild-type littermates. Both male and female Crhr1(-/-)Crhr2(-/-) mutants were viable, fertile, and indistinguishable in size from wild-type littermates. We show that the endocrine phenotype of mice lacking both CRHRs is dominated by the functional loss of CRHR1. CRHR2 does not compensate for CRHR1 deficiency, nor does the lack of CRHR2 exacerbate the CRHR1-dependent impairment of the HPA system function. Within the intraadrenal CRH/ACTH system, our data suggest different roles for CRHR1 and CRHR2 in fine-tuning of adrenocortical corticosterone release.

Molecular characterisation of antidepressant effects in the mouse brain using gene expression profiling.

Antidepressants are widely used for the treatment of psychiatric disorders, including depression and anxiety. Although they are efficient drugs, there are several unsolved questions regarding their clinical pharmacology. Furthermore, the molecular mechanisms of action of antidepressants are still poorly understood and the molecular targets and pathways remain to be identified. To address these issues, we performed a gene expression analysis in mice treated with two commonly used antidepressants with differing pharmacology (paroxetine or mirtazapine) for 1, 7 or 28 days. We quantified the effects of these treatments on gene expression in the mouse brain with cDNA-microarrays containing 3624 expressed sequence tags (ESTs) representing murine genes expressed in the brain. We found that both drugs led to downregulation of four common genes. In addition, although it was possible to identify common targets for the two drugs, the expression profiles of the drugs differed in a fundamental manner, and the longer the treatment duration, the greater the difference in the profiles. These findings suggest that antidepressants with different pharmacologies can share molecular targets even though the primary pathways at which they act are different.

A review of physical abnormalities in familial adenomatous polyposis.

Familial adenomatous polyposis (FAP) is an autosomal dominant condition wherein multiple polyps may be found in the gastrointestinal tract. Initially referred to as familial polyposis coli, it has become evident that virtually all patients with FAP develop adenomas in the upper gastrointestinal tract and thus the syndrome is now termed familial adenomatous polyposis. The number of associated conditions both malignant and benign has been increasingly recognized. Some of these lesions cause morbidity and mortality in affected individuals whilst others act as important clinical markers for identifying patients not yet expressing the phenotype. These abnormalities can arise from tissues of all three primary embryonic layers and are described in this paper.

Anatomic and magnetic resonance imaging bases for the naso-maxillo-cheek flap technique.

A transfacial approach to the deep cranio-maxillo-facial areas by the naso-maxillo-cheek flap technique (NMCF) is indicated for the treatment of some bulky tumors of the naso-pharynx. The procedure requires precise preoperative imaging. This study presents the morphologic bases of this surgical access and the reasonable limits of the excision preoperatively determined by magnetic resonance imaging (MRI). 18 facial and skull specimens were submitted to surgical facial dismantling by the NMCF technique according to Curioni's method. The clinical application in a 66-year-old patient suffering from a neuroblastoma of the olfactory nerve extended into the naso-pharynx is presented. Pre- and postoperative MRI correlations were made in transverse, sagittal and frontal acquisitions. Several structures were preserved in the procedure: facial reliefs, inferior orbital rim and orbital floor, posterior wall of the maxillary sinus covering the pterygopalatine fossa, lateral and medial pterygoid plates and pterygopalatine ganglion with its branches, lateral facial neurovascular pedicle, teeth and soft palate. Other structures were sacrificed: arteries and nerves located at the sites of skin and mucosal incision, and at the sites of osteotomies, ie the infraorbital nerve, the distal part of the greater palatine nerve, the nerves supplying the naso-pharynx, the nasal septum and the nasal conchae, nasolacrimal groove and lacrimal canal. The NMCF technique gives wide access to the deep nasal and nasopharyngeal areas. It is essential to preserve the lateral facial neurovascular pedicle to prevent necrosis of the midface structures. Preservation of the bony architecture surrounding the osteotomy sites is of great importance to allow precise final bone reassembly. Preoperative MRI appears of paramount importance to determine the borders of the lesion and the possibility of block resection.

Biliary structures lead to tumour recurrences after laser-induced interstitial thermotherapy.

BACKGROUND AND OBJECTIVE: Thermal diffusion during laser-induced interstitial thermotherapy (LITT) has not yet been fully investigated in heterogeneous tissue architecture such as liver. LITT was performed on rabbit liver tumours to analyse the role of biliary structures in thermal diffusion. STUDY DESIGN/MATERIALS AND METHODS: Twenty-four VX2 tumours were grafted onto 12 rabbit livers. The animals were randomly separated into two groups when tumour size reached 8 mm. Thermotherapy was performed by delivering the 830-nm output of a diode laser to the centre of the tumour with a 300-,microm fibre. Irradiation conditions were 1.5 W over 900 sec. On day 7 or 14, the tumours were removed and stained with haematoxylin-eosin and picrosirius red F3BA (PR). Thermal damage was evaluated by PR and electron microscopic examinations. RESULTS: Among the treated tumours, recurrences were found both at the periphery (one on day 7, seven on day 14) and within the treated area (two on day 7, two on day 14). All recurrences were located in the vicinity of the biliary structures, which are frequently spared from thermal injury. CONCLUSION: Biliary ducts lead to a heat sink, thereby facilitating tumour recurrences.

Desmoid tumours complicating familial adenomatous polyposis.

BACKGROUND: Desmoid tumours are one of the most important and intriguing extracolonic manifestations of familial adenomatous polyposis (FAP). They have been studied only in small numbers of patients. METHODS: Patients with FAP who also had desmoid tumour were identified from a polyposis registry database and their hospital notes were reviewed. RESULTS: There were 166 desmoids in 88 patients (median age 32 (interquartile range 22-38) years; 51 (58 per cent) female); 83 tumours (50 per cent) were within the abdomen and 80 (48 per cent) were in the abdominal wall. All but 16 individuals (18 per cent) had already undergone abdominal surgery, which was significantly more recent in women (P = 0.01, Mann-Whitney U test). Intra-abdominal desmoids caused small bowel and ureteric obstruction and resulted in ten deaths; survival was significantly poorer than in patients with abdominal wall desmoid alone (chi2 = 3. 93, 1 d.f., P = 0.047, log rank test), and eight of 22 patients who underwent resection of intra-abdominal desmoid died in the perioperative period. CONCLUSION: Abdominal wall desmoids caused no deaths or significant morbidity; although recurrence was common after excision, the treatment was safe. Intra-abdominal desmoids can cause serious complications and treatment is often unsuccessful; in particular, surgery for desmoids at this site is hazardous.

Comparison of morbidity and function after colectomy with ileorectal anastomosis or restorative proctocolectomy for familial adenomatous polyposis.

Restorative proctocolectomy with an ileal reservoir (RPC) should prevent colorectal cancer in patients with familial adenomatous polyposis. Until this is confirmed its role compared with total colectomy and ileorectal anastomosis (IRA) will depend on the relative morbidity and postoperative bowel function after the two procedures. This was analysed in 99 patients (37 RPC, 62 IRA) operated on between 1977 and 1989. Morbidity was greater after RPC with subsequent ileostomy closure (median hospital stay, 24 versus 11 days; complications, 60 versus 21 per cent; reoperation, 29 versus 3 per cent; return to normal activity; 31 versus 14 weeks). There was little difference in bowel function; after IRA median frequency was 3/24 h and urgency (unable to wait 15 min) occurred in 50 per cent, compared with 4.5/24h and 17 per cent after RPC. Night evacuation occurred in 10 and 43 per cent respectively. IRA was performed in younger patients (median 19 versus 31 years) who had fewer bowel motions before operation (2 versus 5/24 h). The greater morbidity of RPC suggests that it should be restricted to patients at higher risk of developing later rectal cancer, including those unavailable for follow-up and those with large or confluent rectal polyps or with curable colon cancer at the initial colectomy.