We propose an integrative self-organizing map (iSOM) for exploring differential expression patterns across multiple microarray experiments. The algorithm is based on the assumption that observed differential expressio...We propose an integrative self-organizing map (iSOM) for exploring differential expression patterns across multiple microarray experiments. The algorithm is based on the assumption that observed differential expressions are random samples of a mean pattern model which is unknowna priori. The learning mechanism of iSOM is similar to the conventional SOM. The mean pattern model which underlies the proposed iSOM models mean differential expressions using a one-dimension of mean differential expressions for the mean differential expressions. The feature map of an iSOM model can be used to reveal correlation between multiple medically/biologically related disease types or multiple platform experiments for one disease. We illustrate applications of iSOM using simulated data and real data.展开更多
Objective: To determine to what extent individuals with various family histories of colorectal cancer (from one to three or more affected first degree relatives) benefit from colonoscopic surveillance. Design: Prospec...Objective: To determine to what extent individuals with various family histories of colorectal cancer (from one to three or more affected first degree relatives) benefit from colonoscopic surveillance. Design: Prospective, observational study of high risk families, followed up over 16 years. Setting: Tertiary referral family cancer clinic in London. Participants: 1678 individuals from families registered with die clinic. Individuals were classified according to die strength of their family history: hereditary non-polyposis colorectal cancer (if they fulfilled the Amsterdam criteria), and one, two, or three affected first degree relatives (moderate risk). Interventions: Colonoscopy was initially offered at five year intervals or three year intervals if an adenoma was detected. Main outcome measures: The incidence of adenomas with high risk pathological features or cancer. This was analysed by age, the extent of the family history, and findings on previous colonoscopies. The cohort was flagged for cancer and death. Incidence of colorectal cancer and mortality during over 15 000 person years of follow-up were compared with those expected in the absence of surveillance. Results: High risk adenomas and cancer were most common in families with hereditary non-polyposis colorectal cancer (on initial colonoscopy 5.7%and 0.9%, respectively). In the families with moderate risk, these findings were particularly uncommon under age 45 (1.1%and 0%) and on follow-up colonoscopy if advanced neoplasia was absent initially (1.7%and 0.1%). The incidence of colorectal cancer was substantially lower-80%in families with moderate risk (P = 0.00004), and 43%in families with hereditary non-polyposis colorectal cancer (P = 0.06) than the expected incidence in the absence of surveillance when the family history was taken into account. Conclusions: Colonoscopic surveillance reduces the risk of colorectal cancer in people with a strong family history. This study confirms that members of families with hereditary non-polyposis colorectal cancer require surveillance with short intervals. Individuals with a lesser family history may not require surveillance under age 45, and if advanced neoplasia is absent on initial colonoscopy, surveillance intervals may be lengthened. This would reduce the demand for colonoscopic surveillance.展开更多
文摘We propose an integrative self-organizing map (iSOM) for exploring differential expression patterns across multiple microarray experiments. The algorithm is based on the assumption that observed differential expressions are random samples of a mean pattern model which is unknowna priori. The learning mechanism of iSOM is similar to the conventional SOM. The mean pattern model which underlies the proposed iSOM models mean differential expressions using a one-dimension of mean differential expressions for the mean differential expressions. The feature map of an iSOM model can be used to reveal correlation between multiple medically/biologically related disease types or multiple platform experiments for one disease. We illustrate applications of iSOM using simulated data and real data.
文摘Objective: To determine to what extent individuals with various family histories of colorectal cancer (from one to three or more affected first degree relatives) benefit from colonoscopic surveillance. Design: Prospective, observational study of high risk families, followed up over 16 years. Setting: Tertiary referral family cancer clinic in London. Participants: 1678 individuals from families registered with die clinic. Individuals were classified according to die strength of their family history: hereditary non-polyposis colorectal cancer (if they fulfilled the Amsterdam criteria), and one, two, or three affected first degree relatives (moderate risk). Interventions: Colonoscopy was initially offered at five year intervals or three year intervals if an adenoma was detected. Main outcome measures: The incidence of adenomas with high risk pathological features or cancer. This was analysed by age, the extent of the family history, and findings on previous colonoscopies. The cohort was flagged for cancer and death. Incidence of colorectal cancer and mortality during over 15 000 person years of follow-up were compared with those expected in the absence of surveillance. Results: High risk adenomas and cancer were most common in families with hereditary non-polyposis colorectal cancer (on initial colonoscopy 5.7%and 0.9%, respectively). In the families with moderate risk, these findings were particularly uncommon under age 45 (1.1%and 0%) and on follow-up colonoscopy if advanced neoplasia was absent initially (1.7%and 0.1%). The incidence of colorectal cancer was substantially lower-80%in families with moderate risk (P = 0.00004), and 43%in families with hereditary non-polyposis colorectal cancer (P = 0.06) than the expected incidence in the absence of surveillance when the family history was taken into account. Conclusions: Colonoscopic surveillance reduces the risk of colorectal cancer in people with a strong family history. This study confirms that members of families with hereditary non-polyposis colorectal cancer require surveillance with short intervals. Individuals with a lesser family history may not require surveillance under age 45, and if advanced neoplasia is absent on initial colonoscopy, surveillance intervals may be lengthened. This would reduce the demand for colonoscopic surveillance.
