SEARCH · PubMed Health

Results for “generative artificial intelligence”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,117 records · Page 62Linked to original sources

PubMed2001

The use of a computer-based decision support system facilitates primary care physicians' management of chronic pain.

We tested whether computer-based decision support (CBDS) could enhance the ability of primary care physicians (PCPs) to manage chronic pain. Structured summaries were generated for 50 chronic pain patients referred by PCPs to a pain clinic. A pain specialist used a decision support system to determine appropriate pain therapy and sent letters to the referring physicians outlining these recommendations. Separately, five board-certified PCPs used a CBDS system to "treat" the 50 cases. A successful outcome was defined as one in which new or adjusted therapies recommended by the software were acceptable to the PCPs (i.e., they would have prescribed it to the patient in actual practice). Two pain specialists reviewed the PCPs' outcomes and assigned medical appropriateness scores (0 = totally inappropriate to 10 = totally appropriate). One year later, the hospital database provided information on how the actual patients' pain was managed and the number of patients re-referred by their PCP to the pain clinic. On the basis of CBDS recommendations, the PCP subjects "prescribed" additional pain therapy in 213 of 250 evaluations (85%), with a medical appropriateness score of 5.5 +/- 0.1. Only 25% of these chronic pain patients were subsequently re-referred to the pain clinic within 1 yr. The use of a CBDS system may improve the ability of PCPs to manage chronic pain and may also facilitate screening of consults to optimize specialist utilization.

Adult↗

PubMed2005

Automated detection of prostatic adenocarcinoma from high-resolution ex vivo MRI.

Prostatic adenocarcinoma is the most commonly occurring cancer among men in the United States, second only to skin cancer. Currently, the only definitive method to ascertain the presence of prostatic cancer is by trans-rectal ultrasound (TRUS) directed biopsy. Owing to the poor image quality of ultrasound, the accuracy of TRUS is only 20%-25%. High-resolution magnetic resonance imaging (MRI) has been shown to have a higher accuracy of prostate cancer detection compared to ultrasound. Consequently, several researchers have been exploring the use of high resolution MRI in performing prostate biopsies. Visual detection of prostate cancer, however, continues to be difficult owing to its apparent lack of shape, and the fact that several malignant and benign structures have overlapping intensity and texture characteristics. In this paper, we present a fully automated computer-aided detection (CAD) system for detecting prostatic adenocarcinoma from 4 Tesla ex vivo magnetic resonance (MR) imagery of the prostate. After the acquired MR images have been corrected for background inhomogeneity and nonstandardness, novel three-dimensional (3-D) texture features are extracted from the 3-D MRI scene. A Bayesian classifier then assigns each image voxel a "likelihood" of malignancy for each feature independently. The "likelihood" images generated in this fashion are then combined using an optimally weighted feature combination scheme. Quantitative evaluation was performed by comparing the CAD results with the manually ascertained ground truth for the tumor on the MRI. The tumor labels on the MR slices were determined manually by an expert by visually registering the MR slices with the corresponding regions on the histology slices. We evaluated our CAD system on a total of 33 two-dimensional (2-D) MR slices from five different 3-D MR prostate studies. Five slices from two different glands were used for training. Our feature combination scheme was found to outperform the individual texture features, and also other popularly used feature combination methods, including AdaBoost, ensemble averaging, and majority voting. Further, in several instances our CAD system performed better than the experts in terms of accuracy, the expert segmentations being determined solely from visual inspection of the MRI data. In addition, the intrasystem variability (changes in CAD accuracy with changes in values of system parameters) was significantly lower than the corresponding intraobserver and interobserver variability. CAD performance was found to be very similar for different training sets. Future work will focus on extending the methodology to guide high-resolution MRI-assisted in vivo prostate biopsies.

Adenocarcinoma↗

PubMed2005

Computing spike directivity with tetrodes.

The ability of neurons to generate electrical signals is strongly dependent on the evolution of ion-specific pumps and channels that allow the transfer of charges under the influence of electric fields and concentration gradients. This paper presents a novel method by which flow of these charge fluxes may be computed to provide directivity of charge movement. Simulations of charge flow as well as actual electrophysiological data recorded by tetrodes are used to demonstrate the method. The propagation of charge fluxes in space in data from simulation and actual recordings during action potential can be analyzed using signals recorded by tetrodes. Variation in spike directivity can be estimated by computing singular value decomposition of the estimated 3D trajectory data. The analysis of the spike model can be accomplished by performing simulations of presumed equivalent moving charges recorded by the tetrode tips. For in vivo spike recordings, the variation of spike directivity could be obtained using several spikes of selected neurons considering the charge movement model (CMM). The relationship between computer simulation results and tetrode data recordings is examined. The paper concludes by showing that the method for calculating directivity in actual spike recordings is robust. The method allows for improved filtering of data and more importantly may shed light on furthering the study of spatio-temporal encoding in neurons.

Action Potentials↗

PubMed2003

Measuring the impact of diagnostic decision support on the quality of clinical decision making: development of a reliable and valid composite score.

OBJECTIVE: Few previous studies evaluating the benefits of diagnostic decision support systems have simultaneously measured changes in diagnostic quality and clinical management prompted by use of the system. This report describes a reliable and valid scoring technique to measure the quality of clinical decision plans in an acute medical setting, where diagnostic decision support tools might prove most useful. DESIGN: Sets of differential diagnoses and clinical management plans generated by 71 clinicians for six simulated cases, before and after decision support from a Web-based pediatric differential diagnostic tool (ISABEL), were used. MEASUREMENTS: A composite quality score was calculated separately for each diagnostic and management plan by considering the appropriateness value of each component diagnostic or management suggestion, a weighted sum of individual suggestion ratings, relevance of the entire plan, and its comprehensiveness. The reliability and validity (face, concurrent, construct, and content) of these two final scores were examined. RESULTS: Two hundred fifty-two diagnostic and 350 management suggestions were included in the interrater reliability analysis. There was good agreement between raters (intraclass correlation coefficient, 0.79 for diagnoses, and 0.72 for management). No counterintuitive scores were demonstrated on visual inspection of the sets. Content validity was verified by a consultation process with pediatricians. Both scores discriminated adequately between the plans of consultants and medical students and correlated well with clinicians' subjective opinions of overall plan quality (Spearman rho 0.65, p < 0.01). The diagnostic and management scores for each episode showed moderate correlation (r = 0.51). CONCLUSION: The scores described can be used as key outcome measures in a larger study to fully assess the value of diagnostic decision aids, such as the ISABEL system.

Artificial Intelligence↗

PubMedDate not supplied

Integrating classification trees with local logistic regression in Intensive Care prognosis.

Health care effectiveness and efficiency are under constant scrutiny especially when treatment is quite costly as in the Intensive Care (IC). Currently there are various international quality of care programs for the evaluation of IC. At the heart of such quality of care programs lie prognostic models whose prediction of patient mortality can be used as a norm to which actual mortality is compared. The current generation of prognostic models in IC are statistical parametric models based on logistic regression. Given a description of a patient at admission, these models predict the probability of his or her survival. Typically, this patient description relies on an aggregate variable, called a score, that quantifies the severity of illness of the patient. The use of a parametric model and an aggregate score form adequate means to develop models when data is relatively scarce but it introduces the risk of bias. This paper motivates and suggests a method for studying and improving the performance behavior of current state-of-the-art IC prognostic models. Our method is based on machine learning and statistical ideas and relies on exploiting information that underlies a score variable. In particular, this underlying information is used to construct a classification tree whose nodes denote patient sub-populations. For these sub-populations, local models, most notably logistic regression ones, are developed using only the total score variable. We compare the performance of this hybrid model to that of a traditional global logistic regression model. We show that the hybrid model not only provides more insight into the data but also has a better performance. We pay special attention to the precision aspect of model performance and argue why precision is more important than discrimination ability.

Artificial Intelligence↗

PubMedDate not supplied

The effect of incomplete knowledge on the diagnoses of a computer consultant system.

The knowledge bases (KBs) of diagnostic decision support systems are often incomplete, and gaps in the KB could potentially lead systems to reach diagnoses that are implausible to physicians. To investigate this possibility we studied Iliad (Version 2.01), a computer consultant system that generates differential diagnosis across the domain of internal medicine. Data from the history, physical examination, and laboratory findings of 50 grand-rounds cases were entered into Iliad by a computer consultant aware of the diagnosis but blinded to its presence or absence in Iliad's KB. Two experienced internists were asked to diagnose these cases before and after seeing the results of the computer consultation, and to assess the plausibility of the computer's diagnoses. Twenty-eight of the 50 cases (56.0%) were diseases contained in Iliad's KB. After seeing Iliad's diagnoses for cases in the KB, physicians assigned to their correct diagnoses a higher mean ranked position (1.5 versus 2.0, p less than 0.008) and a higher mean probability (84.0% versus 77.6%, p less than 0.008) compared with their pre-Iliad values, whereas for cases not in the KB, mean position and probability for correct diagnoses did not change. Physician diagnostic accuracy did not change after consultation on cases included or not included in the KB. After adjusting for case difficulty, mean plausibility of Iliad's diagnoses was judged significantly higher (on a seven-point scale) for cases in the KB than for cases not in the KB (4.2 versus 3.2, p less than 0.02).(ABSTRACT TRUNCATED AT 250 WORDS)

Artificial Intelligence↗

PubMed1995

Objects and domains for managing medical data and knowledge.

This paper assesses the object-oriented data paradigm, and describes an algebraic approach which permits the generation of data objects from relational data, based on the knowledge captured in a formal Entity-Relationship model, the Structural Model. The advantage is that now objects can be created that satisfy a variety of particular views, as long as the hierarchies represented by the views are subsumed in the network represented by the overall structural model. The disadvantage of creating view-objects dynamically is that the additional layering has performance implications, so that the speedup expected from object-oriented databases versus relational databases, due to their hierarchical object storage, cannot be realized. However, scalability of systems is increased since large systems tend to have multiple objectives, and hence often multiple valid hierarchical views over the data. This approach has been implemented in the Penguin project, and recently some commercial successors are emerging. In truly large systems new problems arise, namely that now not only multiple views will exist, but also that the domains to be covered by the data will be autonomous and hence heterogeneous. One result is that ontologies associated with the multiple domains will differ as well. This paper proposes a knowledge-based algebra over the ontologies, so that the domain knowledge can be partitioned for maintenance. Only the articulation points, where the domains intersect, have to be agreed upon as defined by matching rules which define the shared ontologies.

Artificial Intelligence↗

PubMed2006

A study of structural and parametric learning in XCS.

The performance of a learning classifier system is due to its two main components. First, it evolves new structures by generating new rules in a genetic process; second, it adjusts parameters of existing rules, for example rule prediction and accuracy, in an evaluation step, which is not only important for applying the rules, but also for the genetic process. The two components interleave and in the case of XCS drive the population toward a minimal, fit, non-overlapping population. In this work we attempt to gain new insights as to the relative contributions of the two components. We find that the genetic component has an additional role when using the train/test approach which is not present in online learning. We compare XCS to a system in which the rule set is restricted to the initial random population (XCS-NGA, that is, XCS No Genetic Algorithm). For small Boolean functions we can give XCS-NGA all possible rules of a particular condition length. In online learning, XCS-NGA can, given sufficiently many rules, achieve a surprisingly high classification accuracy, comparable to that of XCS. In a train/test approach, however, XCS generalises better than XCS-NGA and there seem to be limitations of XCS-NGA which cannot be overcome simply by increasing the population size. This illustrates that the requirements of a function approximator tend to differ between reinforcement learning (which is typically online) and concept learning (which is typically train/test).

Algorithms↗

PubMed1997

An "augmented-reality" aid for plastic and reconstructive surgeons.

Starting from MR and CT images for a given patient, a new single image representation of all parameters has been generated by using false-color techniques in a standard UNIX and X-11 environment. A transformation linking together the MR, CT parameters and the RGB (red, green, blue) color components has been used. Moreover an unsupervised segmentation method based on a number of neural and fuzzy models may directly produce segmented image volumes. Each image of the various sequences has been interactively displayed by using a specifically designed application. The resulting images have been displayed on a stereo monitor allowing the three-dimensional rendering of visual data through LCD shuttered glasses. Moreover, a 3-D control system based on low frequency magnetic fields has been used, while a bandheld Polhemus stylus could be used as an electronic knife for dissecting the 3-D data set and for defining flaps and grafts. Bone or soft-tissue contour can be analyzed, and sections can be removed from the model to allow a view of the underlying structures. Flaps and grafts obtained utilizing the above-reported techniques can be fitted exactly, without repeated removal and recarving. Nuances of depth, tapering, and arc are carved directly into the bone, while chances of asymmetry are markedly diminished. In this way, moreover, anesthetic times are reduced by more efficient utilization of operative time, which usually offsets the increased cost of imaging.

Artificial Intelligence↗

PubMed2006

Rotation forest: A new classifier ensemble method.

We propose a method for generating classifier ensembles based on feature extraction. To create the training data for a base classifier, the feature set is randomly split into K subsets (K is a parameter of the algorithm) and Principal Component Analysis (PCA) is applied to each subset. All principal components are retained in order to preserve the variability information in the data. Thus, K axis rotations take place to form the new features for a base classifier. The idea of the rotation approach is to encourage simultaneously individual accuracy and diversity within the ensemble. Diversity is promoted through the feature extraction for each base classifier. Decision trees were chosen here because they are sensitive to rotation of the feature axes, hence the name "forest." Accuracy is sought by keeping all principal components and also using the whole data set to train each base classifier. Using WEKA, we examined the Rotation Forest ensemble on a random selection of 33 benchmark data sets from the UCI repository and compared it with Bagging, AdaBoost, and Random Forest. The results were favorable to Rotation Forest and prompted an investigation into diversity-accuracy landscape of the ensemble models. Diversity-error diagrams revealed that Rotation Forest ensembles construct individual classifiers which are more accurate than these in AdaBoost and Random Forest, and more diverse than these in Bagging, sometimes more accurate as well.

Algorithms↗

PubMed2004

KASER: Knowledge Amplification by Structured Expert Randomization.

In this paper and attached video, we present a third-generation expert system named Knowledge Amplification by Structured Expert Randomization (KASER) for which a patent has been filed by the U.S. Navy's SPAWAR Systems Center, San Diego, CA (SSC SD). KASER is a creative expert system. It is capable of deductive, inductive, and mixed derivations. Its qualitative creativity is realized by using a tree-search mechanism. The system achieves creative reasoning by using a declarative representation of knowledge consisting of object trees and inheritance. KASER computes with words and phrases. It possesses a capability for metaphor-based explanations. This capability is useful in explaining its creative suggestions and serves to augment the capabilities provided by the explanation subsystems of conventional expert systems. KASER also exhibits an accelerated capability to learn. However, this capability depends on the particulars of the selected application domain. For example, application domains such as the game of chess exhibit a high degree of geometric symmetry. Conversely, application domains such as the game of craps played with two dice exhibit no predictable pattern, unless the dice are loaded. More generally, we say that domains whose informative content can be compressed to a significant degree without loss (or with relatively little loss) are symmetric. Incompressible domains are said to be asymmetric or random. The measure of symmetry plus the measure of randomness must always sum to unity.

Algorithms↗

PubMed2006

Detection and identification of bacteria using antibiotic susceptibility and a multi-array electrochemical sensor with pattern recognition.

This work proposes the use of amperometric signals generated by a 96-well multi-array dissolved oxygen multi-electrode sensor (DOX) coupled with principal component analysis for continuous monitoring, identification and differentiation of bacteria. Two types of differentiation mechanisms were tested: (1) direct monitoring of respiratory activity via oxygen consumption and (2) quantification of the effect of three broad-spectrum antibiotics on bacteria growth and respiration over time. Five species of bacteria were examined including: Escherichia coli, Escherichia adecarboxylata, Comamonas acidovorans, Corynebacterium glutamicum and Staphylococcus epidermidis. The addition of small concentrations of antibiotics to the growth medium alters the oxygen consumption of the cells and a unique fingerprint is created for a specific cell. This fingerprint is shown to evolve over a specific concentration range that is dependant of instrumental constraints of the DOX system. The application of principal component analysis (PCA) to classify the data was also examined. It was shown that bacteria could be classified simply by their oxygen consumption rates over a varying concentration range. Discrimination between species can also be increased by the effects of the antibiotics on the oxygen consumption of varying concentrations of cells. The proposed DOX-PCA system illustrates a generic template that can be tailored to meet specific research goals by the selection of specific cell/antibiotic combinations and concentrations.

Algorithms↗

PubMed2000

Support vector machine classification and validation of cancer tissue samples using microarray expression data.

MOTIVATION: DNA microarray experiments generating thousands of gene expression measurements, are being used to gather information from tissue and cell samples regarding gene expression differences that will be useful in diagnosing disease. We have developed a new method to analyse this kind of data using support vector machines (SVMs). This analysis consists of both classification of the tissue samples, and an exploration of the data for mis-labeled or questionable tissue results. RESULTS: We demonstrate the method in detail on samples consisting of ovarian cancer tissues, normal ovarian tissues, and other normal tissues. The dataset consists of expression experiment results for 97,802 cDNAs for each tissue. As a result of computational analysis, a tissue sample is discovered and confirmed to be wrongly labeled. Upon correction of this mistake and the removal of an outlier, perfect classification of tissues is achieved, but not with high confidence. We identify and analyse a subset of genes from the ovarian dataset whose expression is highly differentiated between the types of tissues. To show robustness of the SVM method, two previously published datasets from other types of tissues or cells are analysed. The results are comparable to those previously obtained. We show that other machine learning methods also perform comparably to the SVM on many of those datasets. AVAILABILITY: The SVM software is available at http://www.cs. columbia.edu/ approximately bgrundy/svm.

Acute Disease↗

PubMed2006

Determinants of ischemic mitral regurgitation in patients with chronic anterior wall myocardial infarction: a real time three-dimensional echocardiography study.

OBJECTIVE: We sought to elucidate the geometric determinants of ischemic mitral regurgitation (IMR) in patients with chronic anterior myocardial infarction (MI). MATERIALS AND METHODS: In 16 patients with anterior MI only (Group A) and 18 patients with both anterior and inferoposterior MI (Group B), three parallel equidistant anteroposterior (AP) planes (medial, central, lateral) perpendicular to the mitral valvular commissure-commissure plane were generated. The systolic tenting area of the mitral valve (MVTa) and the angles between the annular plane and leaflets (anterior, Aalpha; posterior, Palpha) on the AP planes were measured. The left ventricular end-systolic and end-diastolic volumes, and end-diastolic and end-systolic mitral annular area (MAAs) were obtained. RESULT: The regurgitant orifice area (ROA) was significantly smaller in Group A than Group B (0.08 +/- 0.09 vs 0.20 +/- 0.18 cm(2), P < 0.05). In the total of 34 patients, the medial MVTa (P < 0.001), MAAs (P < 0.05) and the spherical index (P < 0.05) were three independent determinants of ROA while the left ventricular volumes were not. MAAs was the only independent determinant of ROA in Group A, while the medial MVTa was in Group B. Palpha (P < 0.05) and MVTa (P = 0.06) tended to be larger in the medial than the lateral side in Group B, while no differences were found in Group A. CONCLUSION: The geometry of the mitral valve apparatus was more important than the left ventricular volumes in determining the severity of IMR in patients with anterior MI. The posteromedial side tenting could play a critical role in causing significant IMR when the inferoposterior MI coexists with anterior MI.

Aged↗

PubMedDate not supplied

EON: a component-based approach to automation of protocol-directed therapy.

Provision of automated support for planning protocol-directed therapy requires a computer program to take as input clinical data stored in an electronic patient-record system and to generate as output recommendations for therapeutic interventions and laboratory testing that are defined by applicable protocols. This paper presents a synthesis of research carried out at Stanford University to model the therapy-planning task and to demonstrate a component-based architecture for building protocol-based decision-support systems. We have constructed general-purpose software components that (1) interpret abstract protocol specifications to construct appropriate patient-specific treatment plans; (2) infer from time-stamped patient data higher-level, interval-based, abstract concepts; (3) perform time-oriented queries on a time-oriented patient database; and (4) allow acquisition and maintenance of protocol knowledge in a manner that facilitates efficient processing both by humans and by computers. We have implemented these components in a computer system known as EON. Each of the components has been developed, evaluated, and reported independently. We have evaluated the integration of the components as a composite architecture by implementing T-HELPER, a computer-based patient-record system that uses EON to offer advice regarding the management of patients who are following clinical trial protocols for AIDS or HIV infection. A test of the reuse of the software components in a different clinical domain demonstrated rapid development of a prototype application to support protocol-based care of patients who have breast cancer.

Algorithms↗

PubMed1994

AIDA: an interactive diabetes advisor.

AIDA is a prototype computer system that incorporates a model of glucose-insulin interaction in type I diabetes mellitus alongside a knowledge-based system to make glycaemic predictions and to generate insulin dosage adjustment advice. The model attempts to reflect the underlying (patho)physiology of insulin action and carbohydrate absorption in quantitative terms. The prototype is intended to be used as a decision support system by clinical personnel in the context of day-to-day management of insulin-dependent diabetic patients. It is designed for use during consultations, as a simulator of patient response following changed insulin and dietary regimen and as a system for providing education on planning insulin therapy. Joe Daniels is a 41-year-old, 70-kg, male insulin-dependent diabetic patient who was diagnosed as being diabetic in 1972, at the age of 22. Joe recently found that he was having hypoglycaemic symptoms. Using self-monitoring blood glucose equipment, glycaemic levels below 3.0 mmol/l were recorded at least once a week, while hyperglycaemic readings (> 16 mmol/l) were observed two to three times per week. Joe came into hospital to have his glycaemic control improved, as doctors were concerned about the risks of him suffering a serious hypoglycaemic attack. Using some of the data collected by Joe while in hospital, we will demonstrate how AIDA might be applied either in a clinical setting to provide therapeutic advice or in an educational setting to interactively teach diabetic patients about their diabetes and educate them to adjust their own insulin injections and diet.

Adult↗

PubMed2005

Smoothing lung segmentation surfaces in three-dimensional X-ray CT images using anatomic guidance.

RATIONALE AND OBJECTIVES: Automatic lung segmentation in volumetric computed tomography (CT) images has been extensively investigated, and several methods have been proposed. Most methods distinguish the lung parenchyma from the surrounding anatomy based on the difference in CT attenuation values. This leads to an irregular and inconsistent lung boundary for the regions near the mediastinum, which can cause inconsistent boundaries both across subjects and within subjects scanned at different intervals of time. Processes like lung image registration and lung atlas construction can be affected by such inconsistencies. Therefore there is a need for a more consistent lung surface near the mediastinum. MATERIALS AND METHODS: This paper presents a fully automatic method for the three-dimensional smoothing of the lung boundary using information from the segmented human airway tree. First, using the segmented airway tree, we define a bounding box around the mediastinum for each lung, within which all operations are performed. We then define all generations of the airway tree distal to the right and left main stem bronchi to be part of the respective lungs and exclude all other segments. Finally, we perform a fast morphological closing with an ellipsoidal kernel to smooth the surface of the lung. RESULTS: This method has been tested by processing the segmented lungs from eight normal datasets. The mean value of the magnitude of curvedness of the lung contours, averaged over all the datasets, is 0.13 postsmoothing, compared with 4.91 before smoothing. The accuracy of the lung contours after smoothing is assessed by comparing the automatic results to manually traced smooth lung borders by a human analyst. Averaged over all volumes, the root mean square difference between human and computer borders is 0.8691 mm after smoothing, compared with 1.3012 mm before smoothing. The mean similarity index, which is an area overlap measure based on the kappa statistic, is 0.9958 (SD 0.0032), after smoothing. CONCLUSIONS: We have described a novel scheme for smoothing the lung contour around the mediastinum. The method is based on using anatomic information from the segmented airway tree. The validation results show that there is good agreement between manual and computer results. Because there are no accepted criteria for defining the lung boundary near the mediastinum, we believe our method of defining the boundary based on the structure of the airway tree provides a good basis for three-dimensional smoothing.

Algorithms↗

PubMedDate not supplied

Combining sequence and time series expression data to learn transcriptional modules.

Our goal is to cluster genes into transcriptional modules--sets of genes where similarity in expression is explained by common regulatory mechanisms at the transcriptional level. We want to learn modules from both time series gene expression data and genome-wide motif data that are now readily available for organisms such as S. cereviseae as a result of prior computational studies or experimental results. We present a generative probabilistic model for combining regulatory sequence and time series expression data to cluster genes into coherent transcriptional modules. Starting with a set of motifs representing known or putative regulatory elements (transcription factor binding sites) and the counts of occurrences of these motifs in each gene's promoter region, together with a time series expression profile for each gene, the learning algorithm uses expectation maximization to learn module assignments based on both types of data. We also present a technique based on the Jensen-Shannon entropy contributions of motifs in the learned model for associating the most significant motifs to each module. Thus, the algorithm gives a global approach for associating sets of regulatory elements to "modules" of genes with similar time series expression profiles. The model for expression data exploits our prior belief of smooth dependence on time by using statistical splines and is suitable for typical time course data sets with relatively few experiments. Moreover, the model is sufficiently interpretable that we can understand how both sequence data and expression data contribute to the cluster assignments, and how to interpolate between the two data sources. We present experimental results on the yeast cell cycle to validate our method and find that our combined expression and motif clustering algorithm discovers modules with both coherent expression and similar motif patterns, including binding motifs associated to known cell cycle transcription factors.

Algorithms↗