Gene expression data from microarrays are being applied to predict preclinical and clinical endpoints, but the reliability of these predictions has not been established. In the MAQC-II project, 36 independent teams analyzed six microarray data sets to generate predictive models for classifying a sample with respect to one of 13 endpoints indicative of lung or liver toxicity in rodents, or of breast cancer, multiple myeloma or neuroblastoma in humans. In total, >30,000 models were built using many combinations of analytical methods. The teams generated predictive models without knowing the biological meaning of some of the endpoints and, to mimic clinical reality, tested the models on data that had not been used for training. We found that model performance depended largely on the endpoint and team proficiency and that different approaches generated models of similar performance. The conclusions and recommendations from MAQC-II should be useful for regulatory agencies, study committees and independent investigators that evaluate methods for global gene expression analysis.
At a glance
As part of the United States Food and Drug Administration's (FDA's) Critical Path Initiative to medical product development (http://www.fda.gov/oc/initiatives/criticalpath/), the MAQC consortium began in February 2005 with the goal of addressing various microarray reliability concerns raised in publications1, 2, 3, 4, 5, 6, 7, 8, 9 pertaining to reproducibility of gene signatures. The first phase of this project (MAQC-I) extensively evaluated the technical performance of microarray platforms in identifying all differentially expressed genes that would potentially constitute biomarkers. The MAQC-I found high intra-platform reproducibility across test sites, as well as inter-platform concordance of differentially expressed gene lists10, 11, 12, 13, 14, 15 and confirmed that microarray technology is able to reliably identify differentially expressed genes between sample classes or populations16, 17. Importantly, the MAQC-I helped produce companion guidance regarding genomic data submission to the FDA (http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/ucm079855.pdf).
Although the MAQC-I focused on the technical aspects of gene expression measurements, robust technology platforms alone are not sufficient to fully realize the promise of this technology. An additional requirement is the development of accurate and reproducible multivariate gene expression–based prediction models, also referred to as classifiers. Such models take gene expression data from a patient as input and as output produce a prediction of a clinically relevant outcome for that patient. Therefore, the second phase of the project (MAQC-II) has focused on these predictive models18, studying both how they are developed and how they are evaluated. For any given microarray data set, many computational approaches can be followed to develop predictive models and to estimate the future performance of these models. Understanding the strengths and limitations of these various approaches is critical to the formulation of guidelines for safe and effective use of preclinical and clinical genomic data. Although previous studies have compared and benchmarked individual steps in the model development process19, no prior published work has, to our knowledge, extensively evaluated current community practices on the development and validation of microarray-based predictive models.
Microarray-based gene expression data and prediction models are increasingly being submitted by the regulated industry to the FDA to support medical product development and testing applications20. For example, gene expression microarray–based assays that have been approved by the FDA as diagnostic tests include the Agendia MammaPrint microarray to assess prognosis of distant metastasis in breast cancer patients21, 22 and the Pathwork Tissue of Origin Test to assess the degree of similarity of the RNA expression pattern in a patient's tumor to that in a database of tumor samples for which the origin of the tumor is known23. Gene expression data have also been the basis for the development of PCR-based diagnostic assays, including the xDx Allomap test for detection of rejection of heart transplants24.
The possible uses of gene expression data are vast and include diagnosis, early detection (screening), monitoring of disease progression, risk assessment, prognosis, complex medical product characterization and prediction of response to treatment (with regard to safety or efficacy) with a drug or device labeling intent. The ability to generate models in a reproducible fashion is an important consideration in predictive model development.
A lack of consistency in generating classifiers from publicly available data is problematic and may be due to any number of factors including insufficient annotation, incomplete clinical identifiers, coding errors and/or inappropriate use of methodology25, 26. There are also examples in the literature of classifiers whose performance cannot be reproduced on independent data sets because of poor study design27, poor data quality and/or insufficient cross-validation of all model development steps28, 29. Each of these factors may contribute to a certain level of skepticism about claims of performance levels achieved by microarray-based classifiers.
Previous evaluations of the reproducibility of microarray-based classifiers, with only very few exceptions30, 31, have been limited to simulation studies or reanalysis of previously published results. Frequently, published benchmarking studies have split data sets at random, and used one part for training and the other for validation. This design assumes that the training and validation sets are produced by unbiased sampling of a large, homogeneous population of samples. However, specimens in clinical studies are usually accrued over years and there may be a shift in the participating patient population and also in the methods used to assign disease status owing to changing practice standards. There may also be batch effects owing to time variations in tissue analysis or due to distinct methods of sample collection and handling at different medical centers. As a result, samples derived from sequentially accrued patient populations, as was done in MAQC-II to mimic clinical reality, where the first cohort is used for developing predictive models and subsequent patients are included in validation, may differ from each other in many ways that could influence the prediction performance.
The MAQC-II project was designed to evaluate these sources of bias in study design by constructing training and validation sets at different times, swapping the test and training sets and also using data from diverse preclinical and clinical scenarios. The goals of MAQC-II were to survey approaches in genomic model development in an attempt to understand sources of variability in prediction performance and to assess the influences of endpoint signal strength in data. By providing the same data sets to many organizations for analysis, but not restricting their data analysis protocols, the project has made it possible to evaluate to what extent, if any, results depend on the team that performs the analysis. This contrasts with previous benchmarking studies that have typically been conducted by single laboratories. Enrolling a large number of organizations has also made it feasible to test many more approaches than would be practical for any single team. MAQC-II also strives to develop good modeling practice guidelines, drawing on a large international collaboration of experts and the lessons learned in the perhaps unprecedented effort of developing and evaluating >30,000 genomic classifiers to predict a variety of endpoints from diverse data sets.
MAQC-II is a collaborative research project that includes participants from the FDA, other government agencies, industry and academia. This paper describes the MAQC-II structure and experimental design and summarizes the main findings and key results of the consortium, whose members have learned a great deal during the process. The resulting guidelines are general and should not be construed as specific recommendations by the FDA for regulatory submissions.
Generating a unique compendium of >30,000 prediction models
The MAQC-II consortium was conceived with the primary goal of examining model development practices for generating binary classifiers in two types of data sets, preclinical and clinical (Supplementary Tables 1 and 2). To accomplish this, the project leader distributed six data sets containing 13 preclinical and clinical endpoints coded A through M (Table 1) to 36 voluntary participating data analysis teams representing academia, industry and government institutions (Supplementary Table 3). Endpoints were coded so as to hide the identities of two negative-control endpoints (endpoints I and M, for which class labels were randomly assigned and are not predictable by the microarray data) and two positive-control endpoints (endpoints H and L, representing the sex of patients, which is highly predictable by the microarray data). Endpoints A, B and C tested teams' ability to predict the toxicity of chemical agents in rodent lung and liver models. The remaining endpoints were predicted from microarray data sets from human patients diagnosed with breast cancer (D and E), multiple myeloma (F and G) or neuroblastoma (J and K). For the multiple myeloma and neuroblastoma data sets, the endpoints represented event free survival (abbreviated EFS), meaning a lack of malignancy or disease recurrence, and overall survival (abbreviated OS) after 730 days (for multiple myeloma) or 900 days (for neuroblastoma) post treatment or diagnosis. For breast cancer, the endpoints represented estrogen receptor status, a common diagnostic marker of this cancer type (abbreviated 'erpos'), and the success of treatment involving chemotherapy followed by surgical resection of a tumor (abbreviated 'pCR'). The biological meaning of the control endpoints was known only to the project leader and not revealed to the project participants until all model development and external validation processes had been completed.
To evaluate the reproducibility of the models developed by a data analysis team for a given data set, we asked teams to submit models from two stages of analyses. In the first stage (hereafter referred to as the 'original' experiment), each team built prediction models for up to 13 different coded endpoints using six training data sets. Models were 'frozen' against further modification, submitted to the consortium and then tested on a blinded validation data set that was not available to the analysis teams during training. In the second stage (referred to as the 'swap' experiment), teams repeated the model building and validation process by training models on the original validation set and validating them using the original training set.
To simulate the potential decision-making process for evaluating a microarray-based classifier, we established a process for each group to receive training data with coded endpoints, propose a data analysis protocol (DAP) based on exploratory analysis, receive feedback on the protocol and then perform the analysis and validation (Fig. 1). Analysis protocols were reviewed internally by other MAQC-II participants (at least two reviewers per protocol) and by members of the MAQC-II Regulatory Biostatistics Working Group (RBWG), a team from the FDA and industry comprising biostatisticians and others with extensive model building expertise. Teams were encouraged to revise their protocols to incorporate feedback from reviewers, but each team was eventually considered responsible for its own analysis protocol and incorporating reviewers' feedback was not mandatory (see Online Methods for more details).
We assembled two large tables from the original and swap experiments (Supplementary Tables 1 and 2, respectively) containing summary information about the algorithms and analytic steps, or 'modeling factors', used to construct each model and the 'internal' and 'external' performance of each model. Internal performance measures the ability of the model to classify the training samples, based on cross-validation exercises. External performance measures the ability of the model to classify the blinded independent validation data. We considered several performance metrics, including Matthews Correlation Coefficient (MCC), accuracy, sensitivity, specificity, area under the receiver operating characteristic curve (AUC) and root mean squared error (r.m.s.e.). These two tables contain data on >30,000 models. Here we report performance based on MCC because it is informative when the distribution of the two classes in a data set is highly skewed and because it is simple to calculate and was available for all models. MCC values range from +1 to −1, with +1 indicating perfect prediction (that is, all samples classified correctly and none incorrectly), 0 indicates random prediction and −1 indicating perfect inverse prediction.
The 36 analysis teams applied many different options under each modeling factor for developing models (Supplementary Table 4) including 17 summary and normalization methods, nine batch-effect removal methods, 33 feature selection methods (between 1 and >1,000 features), 24 classification algorithms and six internal validation methods. Such diversity suggests the community's common practices are well represented. For each of the models nominated by a team as being the best model for a particular endpoint, we compiled the list of features used for both the original and swap experiments (see the MAQC Web site at http://edkb.fda.gov/MAQC/). These comprehensive tables represent a unique resource. The results that follow describe data mining efforts to determine the potential and limitations of current practices for developing and validating gene expression–based prediction models.
Performance depends on endpoint and can be estimated during training
Unlike many previous efforts, the study design of MAQC-II provided the opportunity to assess the performance of many different modeling approaches on a clinically realistic blinded external validation data set. This is especially important in light of the intended clinical or preclinical uses of classifiers that are constructed using initial data sets and validated for regulatory approval and then are expected to accurately predict samples collected under diverse conditions perhaps months or years later. To assess the reliability of performance estimates derived during model training, we compared the performance on the internal training data set with performance on the external validation data set for of each of the 18,060 models in the original experiment (Fig. 2a). Models without complete metadata were not included in the analysis.
We selected 13 'candidate models', representing the best model for each endpoint, before external validation was performed. We required that each analysis team nominate one model for each endpoint they analyzed and we then selected one candidate from these nominations for each endpoint. We observed a higher correlation between internal and external performance estimates in terms of MCC for the selected candidate models (r = 0.951, n = 13, Fig. 2b) than for the overall set of models (r = 0.840, n = 18,060, Fig. 2a), suggesting that extensive peer review of analysis protocols was able to avoid selecting models that could result in less reliable predictions in external validation. Yet, even for the hand-selected candidate models, there is noticeable bias in the performance estimated from internal validation. That is, the internal validation performance is higher than the external validation performance for most endpoints (Fig. 2b). However, for some endpoints and for some model building methods or teams, internal and external performance correlations were more modest as described in the following sections.
To evaluate whether some endpoints might be more predictable than others and to calibrate performance against the positive- and negative-control endpoints, we assessed all models generated for each endpoint (Fig. 2c). We observed a clear dependence of prediction performance on endpoint. For example, endpoints C (liver necrosis score of rats treated with hepatotoxicants), E (estrogen receptor status of breast cancer patients), and H and L (sex of the multiple myeloma and neuroblastoma patients, respectively) were the easiest to predict (mean MCC > 0.7). Toxicological endpoints A and B and disease progression endpoints D, F, G, J and K were more difficult to predict (mean MCC ~0.1–0.4). Negative-control endpoints I and M were totally unpredictable (mean MCC ~0), as expected. For 11 endpoints (excluding the negative controls), a large proportion of the submitted models predicted the endpoint significantly better than chance (MCC > 0) and for a given endpoint many models performed similarly well on both internal and external validation (see the distribution of MCC in Fig. 2c). On the other hand, not all the submitted models performed equally well for any given endpoint. Some models performed no better than chance, even for some of the easy-to-predict endpoints, suggesting that additional factors were responsible for differences in model performance.
Data analysis teams show different proficiency
Next, we summarized the external validation performance of the models nominated by the 17 teams that analyzed all 13 endpoints (Fig. 3). Nominated models represent a team's best assessment of its model-building effort. The mean external validation MCC per team over 11 endpoints, excluding negative controls I and M, varied from 0.532 for data analysis team (DAT)24 to 0.263 for DAT3, indicating appreciable differences in performance of the models developed by different teams for the same data. Similar trends were observed when AUC was used as the performance metric (Supplementary Table 5) or when the original training and validation sets were swapped (Supplementary Tables 6 and 7). Table 2 summarizes the modeling approaches that were used by two or more MAQC-II data analysis teams.
Many factors may have played a role in the difference of external validation performance between teams. For instance, teams used different modeling factors, criteria for selecting the nominated models, and software packages and code. Moreover, some teams may have been more proficient at microarray data modeling and better at guarding against clerical errors. We noticed substantial variations in performance among the many K-nearest neighbor algorithm (KNN)-based models developed by four analysis teams (Supplementary Fig. 1). Follow-up investigations identified a few possible causes leading to the discrepancies in performance32. For example, DAT20 fixed the parameter 'number of neighbors' K = 3 in its data analysis protocol for all endpoints, whereas DAT18 varied K from 3 to 15 with a step size of 2. This investigation also revealed that even a detailed but standardized description of model building requested from all groups failed to capture many important tuning variables in the process. The subtle modeling differences not captured may have contributed to the differing performance levels achieved by the data analysis teams. The differences in performance for the models developed by various data analysis teams can also be observed from the changing patterns of internal and external validation performance across the 13 endpoints (Fig. 3, Supplementary Tables 5–7 and Supplementary Figs. 2–4). Our observations highlight the importance of good modeling practice in developing and validating microarray-based predictive models including reporting of computational details for results to be replicated26. In light of the MAQC-II experience, recording structured information about the steps and parameters of an analysis process seems highly desirable to facilitate peer review and reanalysis of results.
Swap and original analyses lead to consistent results
To evaluate the reproducibility of the models generated by each team, we correlated the performance of each team's models on the original training data set to performance on the validation data set and repeated this calculation for the swap experiment (Fig. 4). The correlation varied from 0.698–0.966 on the original experiment and from 0.443–0.954 on the swap experiment. For all but three teams (DAT3, DAT10 and DAT11) the original and swap correlations were within ±0.2, and all but three others (DAT4, DAT13 and DAT36) were within ±0.1, suggesting that the model building process was relatively robust, at least with respect to generating models with similar performance. For some data analysis teams the internal validation performance drastically overestimated the performance of the same model in predicting the validation data. Examination of some of those models revealed several reasons, including bias in the feature selection and cross-validation process28, findings consistent with what was observed from a recent literature survey33.
Previously, reanalysis of a widely cited single study34 found that the results in the original publication were very fragile—that is, not reproducible if the training and validation sets were swapped35. Our observations, except for DAT3, DAT11 and DAT36 with correlation <0.6, mainly resulting from failure of accurately predicting the positive-control endpoint H in the swap analysis (likely owing to operator errors), do not substantiate such fragility in the currently examined data sets. It is important to emphasize that we repeated the entire model building and evaluation processes during the swap analysis and, therefore, stability applies to the model building process for each data analysis team and not to a particular model or approach. Supplementary Figure 5 provides a more detailed look at the correlation of internal and external validation for each data analysis team and each endpoint for both the original (Supplementary Fig. 5a) and swap (Supplementary Fig. 5d) analyses.
As expected, individual feature lists differed from analysis group to analysis group and between models developed from the original and the swapped data. However, when feature lists were mapped to biological processes, a greater degree of convergence and concordance was observed. This has been proposed previously but has never been demonstrated in a comprehensive manner over many data sets and thousands of models as was done in MAQC-II36.
The effect of modeling factors is modest
To rigorously identify potential sources of variance that explain the variability in external-validation performance (Fig. 2c), we applied random effect modeling (Fig. 5a). We observed that the endpoint itself is by far the dominant source of variability, explaining >65% of the variability in the external validation performance. All other factors explain <8% of the total variance, and the residual variance is ~6%. Among the factors tested, those involving interactions with endpoint have a relatively large effect, in particular the interaction between endpoint with organization and classification algorithm, highlighting variations in proficiency between analysis teams.
To further investigate the impact of individual levels within each modeling factor, we estimated the empirical best linear unbiased predictors (BLUPs)37. Figure 5b shows the plots of BLUPs of the corresponding factors in Figure 5a with proportion of variation >1%. The BLUPs reveal the effect of each level of the factor to the corresponding MCC value. The BLUPs of the main endpoint effect show that rat liver necrosis, breast cancer estrogen receptor status and the sex of the patient (endpoints C, E, H and L) are relatively easier to be predicted with ~0.2–0.4 advantage contributed on the corresponding MCC values. The rest of the endpoints are relatively harder to be predicted with about −0.1 to −0.2 disadvantage contributed to the corresponding MCC values. The main factors of normalization, classification algorithm, the number of selected features and the feature selection method have an impact of −0.1 to 0.1 on the corresponding MCC values. Loess normalization was applied to the endpoints (J, K and L) for the neuroblastoma data set with the two-color Agilent platform and has 0.1 advantage to MCC values. Among the Microarray Analysis Suite version 5 (MAS5), Robust Multichip Analysis (RMA) and dChip normalization methods that were applied to all endpoints (A, C, D, E, F, G and H) for Affymetrix data, the dChip method has a lower BLUP than the others. Because normalization methods are partially confounded with endpoints, it may not be suitable to compare methods between different confounded groups. Among classification methods, discriminant analysis has the largest positive impact of 0.056 on the MCC values. Regarding the number of selected features, larger bin number has better impact on the average across endpoints. The bin number is assigned by applying the ceiling function to the log base 10 of the number of selected features. All the feature selection methods have a slight impact of −0.025 to 0.025 on MCC values except for recursive feature elimination (RFE) that has an impact of −0.006. In the plots of the four selected interactions, the estimated BLUPs vary across endpoints. The large variation across endpoints implies the impact of the corresponding modeling factor on different endpoints can be very different. Among the four interaction plots (see Supplementary Fig. 6 for a clear labeling of each interaction term), the corresponding BLUPs of the three-way interaction of organization, classification algorithm and endpoint show the highest variation. This may be due to different tuning parameters applied to individual algorithms for different organizations, as was the case for KNN32.
We also analyzed the relative importance of modeling factors on external-validation prediction performance using a decision tree model38. The analysis results revealed observations (Supplementary Fig. 7) largely consistent with those above. First, the endpoint code was the most influential modeling factor. Second, feature selection method, normalization and summarization method, classification method and organization code also contributed to prediction performance, but their contribution was relatively small.
Feature list stability is correlated with endpoint predictability
Prediction performance is the most important criterion for evaluating the performance of a predictive model and its modeling process. However, the robustness and mechanistic relevance of the model and the corresponding gene signature is also important (Supplementary Fig. 8). That is, given comparable prediction performance between two modeling processes, the one yielding a more robust and reproducible gene signature across similar data sets (e.g., by swapping the training and validation sets), which is therefore less susceptible to sporadic fluctuations in the data, or the one that provides new insights to the underlying biology is preferable. Reproducibility or stability of feature sets is best studied by running the same model selection protocol on two distinct collections of samples, a scenario only possible, in this case, after the blind validation data were distributed to the data analysis teams that were asked to perform their analysis after swapping their original training and test sets. Supplementary Figures 9 and 10 show that, although the feature space is extremely large for microarray data, different teams and protocols were able to consistently select the best-performing features. Analysis of the lists of features indicated that for endpoints relatively easy to predict, various data analysis teams arrived at models that used more common features and the overlap of the lists from the original and swap analyses is greater than those for more difficult endpoints (Supplementary Figs. 9–11). Therefore, the level of stability of feature lists can be associated to the level of difficulty of the prediction problem (Supplementary Fig. 11), although multiple models with different feature lists and comparable performance can be found from the same data set39. Functional analysis of the most frequently selected genes by all data analysis protocols shows that many of these genes represent biological processes that are highly relevant to the clinical outcome that is being predicted36. The sex-based endpoints have the best overlap, whereas more difficult survival endpoints (in which disease processes are confounded by many other factors) have only marginally better overlap with biological processes relevant to the disease than that expected by random chance.
Summary of MAQC-II observations and recommendations
The MAQC-II data analysis teams comprised a diverse group, some of whom were experienced microarray analysts whereas others were graduate students with little experience. In aggregate, the group's composition likely mimicked the broad scientific community engaged in building and publishing models derived from microarray data. The more than 30,000 models developed by 36 data analysis teams for 13 endpoints from six diverse clinical and preclinical data sets are a rich source from which to highlight several important observations.
First, model prediction performance was largely endpoint (biology) dependent (Figs. 2c and 3). The incorporation of multiple data sets and endpoints (including positive and negative controls) in the MAQC-II study design made this observation possible. Some endpoints are highly predictive based on the nature of the data, which makes it possible to build good models, provided that sound modeling procedures are used. Other endpoints are inherently difficult to predict regardless of the model development protocol.
Second, there are clear differences in proficiency between data analysis teams (organizations) and such differences are correlated with the level of experience of the team. For example, the top-performing teams shown in Figure 3 were mainly industrial participants with many years of experience in microarray data analysis, whereas bottom-performing teams were mainly less-experienced graduate students or researchers. Based on results from the positive and negative endpoints, we noticed that simple errors were sometimes made, suggesting rushed efforts due to lack of time or unnoticed implementation flaws. This observation strongly suggests that mechanisms are needed to ensure the reliability of results presented to the regulatory agencies, journal editors and the research community. By examining the practices of teams whose models did not perform well, future studies might be able to identify pitfalls to be avoided. Likewise, practices adopted by top-performing teams can provide the basis for developing good modeling practices.
Third, the internal validation performance from well-implemented, unbiased cross-validation shows a high degree of concordance with the external validation performance in a strict blinding process (Fig. 2). This observation was not possible from previously published studies owing to the small number of available endpoints tested in them.
Fourth, many models with similar performance can be developed from a given data set (Fig. 2). Similar prediction performance is attainable when using different modeling algorithms and parameters, and simple data analysis methods often perform as well as more complicated approaches32, 40. Although it is not essential to include the same features in these models to achieve comparable prediction performance, endpoints that were easier to predict generally yielded models with more common features, when analyzed by different teams (Supplementary Fig. 11).
Finally, applying good modeling practices appeared to be more important than the actual choice of a particular algorithm over the others within the same step in the modeling process. This can be seen in the diverse choices of the modeling factors used by teams that produced models that performed well in the blinded validation (Table 2) where modeling factors did not universally contribute to variations in model performance among good performing teams (Fig. 5).
Summarized below are the model building steps recommended to the MAQC-II data analysis teams. These may be applicable to model building practitioners in the general scientific community.
Step one (design). There is no exclusive set of steps and procedures, in the form of a checklist, to be followed by any practitioner for all problems. However, normal good practice on the study design and the ratio of sample size to classifier complexity should be followed. The frequently used options for normalization, feature selection and classification are good starting points (Table 2).
Step two (pilot study or internal validation). This can be accomplished by bootstrap or cross-validation such as the ten repeats of a fivefold cross-validation procedure adopted by most MAQC-II teams. The samples from the pilot study are not replaced for the pivotal study; rather they are augmented to achieve 'appropriate' target size.
Step three (pivotal study or external validation). Many investigators assume that the most conservative approach to a pivotal study is to simply obtain a test set completely independent of the training set(s). However, it is good to keep in mind the exchange34, 35 regarding the fragility of results when the training and validation sets are swapped. Results from further resampling (including simple swapping as in MAQC-II) across the training and validation sets can provide important information about the reliability of the models and the modeling procedures, but the complete separation of the training and validation sets should be maintained41.
Finally, a perennial issue concerns reuse of the independent validation set after modifications to an originally designed and validated data analysis algorithm or protocol. Such a process turns the validation set into part of the design or training set42. Ground rules must be developed for avoiding this approach and penalizing it when it occurs; and practitioners should guard against using it before such ground rules are well established.
MAQC-II conducted a broad observational study of the current community landscape of gene-expression profile–based predictive model development. Microarray gene expression profiling is among the most commonly used analytical tools in biomedical research. Analysis of the high-dimensional data generated by these experiments involves multiple steps and several critical decision points that can profoundly influence the soundness of the results43. An important requirement of a sound internal validation is that it must include feature selection and parameter optimization within each iteration to avoid overly optimistic estimations of prediction performance28, 29, 44. To what extent this information has been disseminated and followed by the scientific community in current microarray analysis remains unknown33. Concerns have been raised that results published by one group of investigators often cannot be confirmed by others even if the same data set is used26. An inability to confirm results may stem from any of several reasons: (i) insufficient information is provided about the methodology that describes which analysis has actually been done; (ii) data preprocessing (normalization, gene filtering and feature selection) is too complicated and insufficiently documented to be reproduced; or (iii) incorrect or biased complex analytical methods26 are performed. A distinct but related concern is that genomic data may yield prediction models that, even if reproducible on the discovery data set, cannot be extrapolated well in independent validation. The MAQC-II project provided a unique opportunity to address some of these concerns.
Notably, we did not place restrictions on the model building methods used by the data analysis teams. Accordingly, they adopted numerous different modeling approaches (Table 2 and Supplementary Table 4). For example, feature selection methods varied widely, from statistical significance tests, to machine learning algorithms, to those more reliant on differences in expression amplitude, to those employing knowledge of putative biological mechanisms associated with the endpoint. Prediction algorithms also varied widely. To make internal validation performance results comparable across teams for different models, we recommended that a model's internal performance was estimated using a ten times repeated fivefold cross-validation, but this recommendation was not strictly followed by all teams, which also allows us to survey internal validation approaches. The diversity of analysis protocols used by the teams is likely to closely resemble that of current research going forward, and in this context mimics reality. In terms of the space of modeling factors explored, MAQC-II is a survey of current practices rather than a randomized, controlled experiment; therefore, care should be taken in interpreting the results. For example, some teams did not analyze all endpoints, causing missing data (models) that may be confounded with other modeling factors.
Overall, the procedure followed to nominate MAQC-II candidate models was quite effective in selecting models that performed reasonably well during validation using independent data sets, although generally the selected models did not do as well in validation as in training. The drop in performance associated with the validation highlights the importance of not relying solely on internal validation performance, and points to the need to subject every classifier to at least one external validation. The selection of the 13 candidate models from many nominated models was achieved through a peer-review collaborative effort of many experts and could be described as slow, tedious and sometimes subjective (e.g., a data analysis team could only contribute one of the 13 candidate models). Even though they were still subject to over-optimism, the internal and external performance estimates of the candidate models were more concordant than those of the overall set of models. Thus the review was productive in identifying characteristics of reliable models.
An important lesson learned through MAQC-II is that it is almost impossible to retrospectively retrieve and document decisions that were made at every step during the feature selection and model development stage. This lack of complete description of the model building process is likely to be a common reason for the inability of different data analysis teams to fully reproduce each other's results32. Therefore, although meticulously documenting the classifier building procedure can be cumbersome, we recommend that all genomic publications include supplementary materials describing the model building and evaluation process in an electronic format. MAQC-II is making available six data sets with 13 endpoints that can be used in the future as a benchmark to verify that software used to implement new approaches performs as expected. Subjecting new software to benchmarks against these data sets could reassure potential users that the software is mature enough to be used for the development of predictive models in new data sets. It would seem advantageous to develop alternative ways to help determine whether specific implementations of modeling approaches and performance evaluation procedures are sound, and to identify procedures to capture this information in public databases.
The findings of the MAQC-II project suggest that when the same data sets are provided to a large number of data analysis teams, many groups can generate similar results even when different model building approaches are followed. This is concordant with studies29, 33 that found that given good quality data and an adequate number of informative features, most classification methods, if properly used, will yield similar predictive performance. This also confirms reports6, 7, 39 on small data sets by individual groups that have suggested that several different feature selection methods and prediction algorithms can yield many models that are distinct, but have statistically similar performance. Taken together, these results provide perspective on the large number of publications in the bioinformatics literature that have examined the various steps of the multivariate prediction model building process and identified elements that are critical for achieving reliable results.
An important and previously underappreciated observation from MAQC-II is that different clinical endpoints represent very different levels of classification difficulty. For some endpoints the currently available data are sufficient to generate robust models, whereas for other endpoints currently available data do not seem to be sufficient to yield highly predictive models. An analysis done as part of the MAQC-II project and that focused on the breast cancer data demonstrates these points in more detail40. It is also important to point out that for some clinically meaningful endpoints studied in the MAQC-II project, gene expression data did not seem to significantly outperform models based on clinical covariates alone, highlighting the challenges in predicting the outcome of patients in a heterogeneous population and the potential need to combine gene expression data with clinical covariates (unpublished data).
The accuracy of the clinical sample annotation information may also play a role in the difficulty to obtain accurate prediction results on validation samples. For example, some samples were misclassified by almost all models (Supplementary Fig. 12). It is true even for some samples within the positive control endpoints H and L, as shown in Supplementary Table 8. Clinical information of neuroblastoma patients for whom the positive control endpoint L was uniformly misclassified were rechecked and the sex of three out of eight cases (NB412, NB504 and NB522) was found to be incorrectly annotated.
The companion MAQC-II papers published elsewhere give more in-depth analyses of specific issues such as the clinical benefits of genomic classifiers (unpublished data), the impact of different modeling factors on prediction performance45, the objective assessment of microarray cross-platform prediction46, cross-tissue prediction47, one-color versus two-color prediction comparison48, functional analysis of gene signatures36 and recommendation of a simple yet robust data analysis protocol based on the KNN32. For example, we systematically compared the classification performance resulting from one- and two-color gene-expression profiles of 478 neuroblastoma samples and found that analyses based on either platform yielded similar classification performance48. This newly generated one-color data set has been used to evaluate the applicability of the KNN-based simple data analysis protocol to future data sets32. In addition, the MAQC-II Genome-Wide Association Working Group assessed the variabilities in genotype calling due to experimental or algorithmic factors49.
In summary, MAQC-II has demonstrated that current methods commonly used to develop and assess multivariate gene-expression based predictors of clinical outcome were used appropriately by most of the analysis teams in this consortium. However, differences in proficiency emerged and this underscores the importance of proper implementation of otherwise robust analytical methods. Observations based on analysis of the MAQC-II data sets may be applicable to other diseases. The MAQC-II data sets are publicly available and are expected to be used by the scientific community as benchmarks to ensure proper modeling practices. The experience with the MAQC-II clinical data sets also reinforces the notion that clinical classification problems represent several different degrees of prediction difficulty that are likely to be associated with whether mRNA abundances measured in a specific data set are informative for the specific prediction problem. We anticipate that including other types of biological data at the DNA, microRNA, protein or metabolite levels will enhance our capability to more accurately predict the clinically relevant endpoints. The good modeling practice guidelines established by MAQC-II and lessons learned from this unprecedented collaboration provide a solid foundation from which other high-dimensional biological data could be more reliably used for the purpose of predictive and personalized medicine.
MAQC-II participants can be grouped into several categories. Data providers are the participants who provided data sets to the consortium. The MAQC-II Regulatory Biostatistics Working Group, whose members included a number of biostatisticians, provided guidance and standard operating procedures for model development and performance estimation. One or more data analysis teams were formed at each organization. Each data analysis team actively analyzed the data sets and produced prediction models. Other participants also contributed to discussion and execution of the project. The 36 data analysis teams listed in Supplementary Table 3 developed data analysis protocols and predictive models for one or more of the 13 endpoints. The teams included more than 100 scientists and engineers with diverse backgrounds in machine learning, statistics, biology, medicine and chemistry, among others. They volunteered tremendous time and effort to conduct the data analysis tasks.
Six data sets including 13 prediction endpoints.
To increase the chance that MAQC-II would reach generalized conclusions, consortium members strongly believed that they needed to study several data sets, each of high quality and sufficient size, which would collectively represent a diverse set of prediction tasks. Accordingly, significant early effort went toward the selection of appropriate data sets. Over ten nominated data sets were reviewed for quality of sample collection and processing consistency, and quality of microarray and clinical data. Six data sets with 13 endpoints were ultimately selected among those nominated during a face-to-face project meeting with extensive deliberations among many participants (Table 1). Importantly, three preclinical (toxicogenomics) and three clinical data sets were selected to test whether baseline practice conclusions could be generalized across these rather disparate experimental types. An important criterion for data set selection was the anticipated support of MAQC-II by the data provider and the commitment to continue experimentation to provide a large external validation test set of comparable size to the training set. The three toxicogenomics data sets would allow the development of predictive models that predict toxicity of compounds in animal models, a prediction task of interest to the pharmaceutical industry, which could use such models to speed up the evaluation of toxicity for new drug candidates. The three clinical data sets were for endpoints associated with three diseases, breast cancer (BR), multiple myeloma (MM) and neuroblastoma (NB). Each clinical data set had more than one endpoint, and together incorporated several types of clinical applications, including treatment outcome and disease prognosis. The MAQC-II predictive modeling was limited to binary classification problems; therefore, continuous endpoint values such as overall survival (OS) and event-free survival (EFS) times were dichotomized using a 'milestone' cutoff of censor data. Prediction endpoints were chosen to span a wide range of prediction difficulty. Two endpoints, H (CPS1) and L (NEP_S), representing the sex of the patients, were used as positive control endpoints, as they are easily predictable by microarrays. Two other endpoints, I (CPR1) and M (NEP_R), representing randomly assigned class labels, were designed to serve as negative control endpoints, as they are not supposed to be predictable. Data analysis teams were not aware of the characteristics of endpoints H, I, L and M until their swap prediction results had been submitted. If a data analysis protocol did not yield models to accurately predict endpoints H and L, or if a data analysis protocol claims to be able to yield models to accurately predict endpoints I and M, something must have gone wrong.
The Hamner data set (endpoint A) was provided by The Hamner Institutes for Health Sciences. The study objective was to apply microarray gene expression data from the lung of female B6C3F1 mice exposed to a 13-week treatment of chemicals to predict increased lung tumor incidence in the 2-year rodent cancer bioassays of the National Toxicology Program50. If successful, the results may form the basis of a more efficient and economical approach for evaluating the carcinogenic activity of chemicals. Microarray analysis was performed using Affymetrix Mouse Genome 430 2.0 arrays on three to four mice per treatment group, and a total of 70 mice were analyzed and used as MAQC-II's training set. Additional data from another set of 88 mice were collected later and provided as MAQC-II's external validation set.
The Iconix data set (endpoint B) was provided by Iconix Biosciences. The study objective was to assess, upon short-term exposure, hepatic tumor induction by nongenotoxic chemicals51, as there are currently no accurate and well-validated short-term tests to identify nongenotoxic hepatic tumorigens, thus necessitating an expensive 2-year rodent bioassay before a risk assessment can begin. The training set consists of hepatic gene expression data from 216 male Sprague-Dawley rats treated for 5 d with one of 76 structurally and mechanistically diverse nongenotoxic hepatocarcinogens and nonhepatocarcinogens. The validation set consists of 201 male Sprague-Dawley rats treated for 5 d with one of 68 structurally and mechanistically diverse nongenotoxic hepatocarcinogens and nonhepatocarcinogens. Gene expression data were generated using the Amersham Codelink Uniset Rat 1 Bioarray (GE HealthCare)52. The separation of the training set and validation set was based on the time when the microarray data were collected; that is, microarrays processed earlier in the study were used as training and those processed later were used as validation.
The NIEHS data set (endpoint C) was provided by the National Institute of Environmental Health Sciences (NIEHS) of the US National Institutes of Health. The study objective was to use microarray gene expression data acquired from the liver of rats exposed to hepatotoxicants to build classifiers for prediction of liver necrosis. The gene expression 'compendium' data set was collected from 418 rats exposed to one of eight compounds (1,2-dichlorobenzene, 1,4-dichlorobenzene, bromobenzene, monocrotaline, N-nitrosomorpholine, thioacetamide, galactosamine and diquat dibromide). All eight compounds were studied using standardized procedures, that is, a common array platform (Affymetrix Rat 230 2.0 microarray), experimental procedures and data retrieving and analysis processes. For details of the experimental design see ref. 53. Briefly, for each compound, four to six male, 12-week-old F344 rats were exposed to a low dose, mid dose(s) and a high dose of the toxicant and sacrificed 6, 24 and 48 h later. At necropsy, liver was harvested for RNA extraction, histopathology and clinical chemistry assessments.
Animal use in the studies was approved by the respective Institutional Animal Use and Care Committees of the data providers and was conducted in accordance with the National Institutes of Health (NIH) guidelines for the care and use of laboratory animals. Animals were housed in fully accredited American Association for Accreditation of Laboratory Animal Care facilities.
The human breast cancer (BR) data set (endpoints D and E) was contributed by the University of Texas M.D. Anderson Cancer Center. Gene expression data from 230 stage I–III breast cancers were generated from fine needle aspiration specimens of newly diagnosed breast cancers before any therapy. The biopsy specimens were collected sequentially during a prospective pharmacogenomic marker discovery study between 2000 and 2008. These specimens represent 70–90% pure neoplastic cells with minimal stromal contamination54. Patients received 6 months of preoperative (neoadjuvant) chemotherapy including paclitaxel (Taxol), 5-fluorouracil, cyclophosphamide and doxorubicin (Adriamycin) followed by surgical resection of the cancer. Response to preoperative chemotherapy was categorized as a pathological complete response (pCR = no residual invasive cancer in the breast or lymph nodes) or residual invasive cancer (RD), and used as endpoint D for prediction. Endpoint E is the clinical estrogen-receptor status as established by immunohistochemistry55. RNA extraction and gene expression profiling were performed in multiple batches over time using Affymetrix U133A microarrays. Genomic analysis of a subset of this sequentially accrued patient population were reported previously56. For each endpoint, the first 130 cases were used as a training set and the next 100 cases were used as an independent validation set.
The multiple myeloma (MM) data set (endpoints F, G, H and I) was contributed by the Myeloma Institute for Research and Therapy at the University of Arkansas for Medical Sciences. Gene expression profiling of highly purified bone marrow plasma cells was performed in newly diagnosed patients with MM57, 58, 59. The training set consisted of 340 cases enrolled in total therapy 2 (TT2) and the validation set comprised 214 patients enrolled in total therapy 3 (TT3)59. Plasma cells were enriched by anti-CD138 immunomagnetic bead selection of mononuclear cell fractions of bone marrow aspirates in a central laboratory. All samples applied to the microarray contained >85% plasma cells as determined by two-color flow cytometry (CD38+ and CD45−/dim) performed after selection. Dichotomized overall survival (OS) and event-free survival (EFS) were determined based on a 2-year milestone cutoff. A gene expression model of high-risk multiple myeloma was developed and validated by the data provider58 and later on validated in three additional independent data sets60, 61, 62.
The neuroblastoma (NB) data set (endpoints J, K, L and M) was contributed by the Children's Hospital of the University of Cologne, Germany. Tumor samples were checked by a pathologist before RNA isolation; only samples with ≥60% tumor content were used and total RNA was isolated from ~50 mg of snap-frozen neuroblastoma tissue obtained before chemotherapeutic treatment. First, 502 preexisting 11 K Agilent dye-flipped, dual-color replicate profiles for 251 patients were provided63. Of these, profiles of 246 neuroblastoma samples passed an independent MAQC-II quality assessment by majority decision and formed the MAQC-II training data set. Subsequently, 514 dye-flipped dual-color 11 K replicate profiles for 256 independent neuroblastoma tumor samples were generated and profiles for 253 samples were selected to form the MAQC-II validation set. Of note, for one patient of the validation set, two different tumor samples were analyzed using both versions of the 2 × 11K microarray (see below). All dual-color gene-expression of the MAQC-II training set were generated using a customized 2 × 11K neuroblastoma-related microarray63. Furthermore, 20 patients of the MAQC-II validation set were also profiled using this microarray. Dual-color profiles of the remaining patients of the MAQC-II validation set were performed using a slightly revised version of the 2 × 11K microarray. This version V2.0 of the array comprised 200 novel oligonucleotide probes whereas 100 oligonucleotide probes of the original design were removed due to consistent low expression values (near background) observed in the training set profiles. These minor modifications of the microarray design resulted in a total of 9,986 probes present on both versions of the 2 × 11K microarray. The experimental protocol did not differ between both sets and gene-expression profiles were performed as described63. Furthermore, single-color gene-expression profiles were generated for 478/499 neuroblastoma samples of the MAQC-II dual-color training and validation sets (training set 244/246; validation set 234/253). For the remaining 21 samples no single-color data were available, due to either shortage of tumor material of these patients (n = 15), poor experimental quality of the generated single-color profiles (n = 5), or correlation of one single-color profile to two different dual-color profiles for the one patient profiled with both versions of the 2 × 11K microarrays (n = 1). Single-color gene-expression profiles were generated using customized 4 × 44K oligonucleotide microarrays produced by Agilent Technologies. These 4 × 44K microarrays included all probes represented by Agilent's Whole Human Genome Oligo Microarray and all probes of the version V2.0 of the 2 × 11K customized microarray that were not present in the former probe set. Labeling and hybridization was performed following the manufacturer's protocol as described48.
Sample annotation information along with clinical co-variates of the patient cohorts is available at the MAQC web site (http://edkb.fda.gov/MAQC/). The institutional review boards of the respective providers of the clinical microarray data sets had approved the research studies, and all subjects had provided written informed consent to both treatment protocols and sample procurement, in accordance with the Declaration of Helsinki.
MAQC-II effort and data analysis procedure.
This section provides details about some of the analysis steps presented in Figure 1. Steps 2–4 in a first round of analysis was conducted where each data analysis team analyzed MAQC-II data sets to generate predictive models and associated performance estimates. After this first round of analysis, most participants attended a consortium meeting where approaches were presented and discussed. The meeting helped members decide on a common performance evaluation protocol, which most data analysis teams agreed to follow to render performance statistics comparable across the consortium. It should be noted that some data analysis teams decided not to follow the recommendations for performance evaluation protocol and used instead an approach of their choosing, resulting in various internal validation approaches in the final results. Data analysis teams were given 2 months to implement the revised analysis protocol (the group recommended using fivefold stratified cross-validation with ten repeats across all endpoints for the internal validation strategy) and submit their final models. The amount of metadata to collect for characterizing the modeling approach used to derive each model was also discussed at the meeting.
For each endpoint, each team was also required to select one of its submitted models as its nominated model. No specific guideline was given and groups could select nominated models according to any objective or subjective criteria. Because the consortium lacked an agreed upon reference performance measure (Supplementary Fig. 13), it was not clear how the nominated models would be evaluated, and data analysis teams ranked models by different measures or combinations of measures. Data analysis teams were encouraged to report a common set of performance measures for each model so that models could be reranked consistently a posteriori. Models trained with the training set were frozen (step 6). MAQC-II selected for each endpoint one model from the up-to 36 nominations as the MAQC-II candidate for validation (step 6).
External validation sets lacking class labels for all endpoints were distributed to the data analysis teams. Each data analysis team used its previously frozen models to make class predictions on the validation data set (step 7). The sample-by-sample prediction results were submitted to MAQC-II by each data analysis team (step 8). Results were used to calculate the external validation performance metrics for each model. Calculations were carried out by three independent groups not involved in developing models, which were provided with validation class labels. Data analysis teams that still had no access to the validation class labels were given an opportunity to correct apparent clerical mistakes in prediction submissions (e.g., inversion of class labels). Class labels were then distributed to enable data analysis teams to check prediction performance metrics and perform in depth analysis of results. A table of performance metrics was assembled from information collected in steps 5 and 8 (step 10, Supplementary Table 1).
To check the consistency of modeling approaches, the original validation and training sets were swapped and steps 4–10 were repeated (step 11). Briefly, each team used the validation class labels and the validation data sets as a training set. Prediction models and evaluation performance were collected by internal and external validation (considering the original training set as a validation set). Data analysis teams were asked to apply the same data analysis protocols that they used for the original 'Blind' Training Validation analysis. Swap analysis results are provided in Supplementary Table 2. It should be noted that during the swap experiment, the data analysis teams inevitably already had access to the class label information for samples in the swap validation set, that is, the original training set.
Model summary information tables.
To enable a systematic comparison of models for each endpoint, a table of information was constructed containing a row for each model from each data analysis team, with columns containing three categories of information: (i) modeling factors that describe the model development process; (ii) performance metrics from internal validation; and (iii) performance metrics from external validation (Fig. 1; step 10).
Each data analysis team was requested to report several modeling factors for each model they generated. These modeling factors are organization code, data set code, endpoint code, summary or normalization method, feature selection method, number of features used in final model, classification algorithm, internal validation protocol, validation iterations (number of repeats of cross-validation or bootstrap sampling) and batch-effect-removal method. A set of valid entries for each modeling factor was distributed to all data analysis teams in advance of model submission, to help consolidate a common vocabulary that would support analysis of the completed information table. It should be noted that since modeling factors are self-reported, two models that share a given modeling factor may still differ in their implementation of the modeling approach described by the modeling factor.
The seven performance metrics for internal validation and external validation are MCC (Matthews Correlation Coefficient), accuracy, sensitivity, specificity, AUC (area under the receiver operating characteristic curve), binary AUC (that is, mean of sensitivity and specificity) and r.m.s.e. For internal validation, s.d. for each performance metric is also included in the table. Missing entries indicate that the data analysis team has not submitted the requested information.
In addition, the lists of features used in the data analysis team's nominated models are recorded as part of the model submission for functional analysis and reproducibility assessment of the feature lists (see the MAQC Web site at http://edkb.fda.gov/MAQC/).
Selection of nominated models by each data analysis team and selection of MAQC-II candidate and backup models by RBWG and the steering committee.
In addition to providing results to generate the model information table, each team nominated a single model for each endpoint as its preferred model for validation, resulting in a total of 323 nominated models, 318 of which were applied to the prediction of the validation sets. These nominated models were peer reviewed, debated and ranked for each endpoint by the RBWG before validation set predictions. The rankings were given to the MAQC-II steering committee, and those members not directly involved in developing models selected a single model for each endpoint, forming the 13 MAQC-II candidate models. If there was sufficient evidence through documentation to establish that the data analysis team had followed the guidelines of good classifier principles for model development outlined in the standard operating procedure (Supplementary Data), then their nominated models were considered as potential candidate models. The nomination and selection of candidate models occurred before the validation data were released. Selection of one candidate model for each endpoint across MAQC-II was performed to reduce multiple selection concerns. This selection process turned out to be highly interesting, time consuming, but worthy, as participants had different viewpoints and criteria in ranking the data analysis protocols and selecting the candidate model for an endpoint. One additional criterion was to select the 13 candidate models in such a way that only one of the 13 models would be selected from the same data analysis team to ensure that a variety of approaches to model development were considered. For each endpoint, a backup model was also selected under the same selection process and criteria as for the candidate models. The 13 candidate models selected by MAQC-II indeed performed well in the validation prediction (Figs. 2c and 3).
- Getting the noise out of gene arrays. Science 306, 630–631 (2004).
- An array of problems. Nat. Rev. Drug Discov. 4, 362–363 (2005).
- Prediction of cancer outcome with microarrays: a multiple random validation strategy. Lancet 365, 488–492 (2005). , &
- Predictive ability of DNA microarrays for cancer outcomes and correlates: an empirical assessment. Lancet 362, 1439–1444 (2003). &
- Microarrays and molecular research: noise discovery? Lancet 365, 454–455 (2005).
- Outcome signature genes in breast cancer: is there a unique set? Bioinformatics 21, 171–178 (2005). , , , &
- Thousands of samples are needed to generate a robust gene list for predicting outcome in cancer. Proc. Natl. Acad. Sci. USA 103, 5923–5928 (2006). , &
- QA/QC: challenges and pitfalls facing the microarray community and regulatory agencies. Expert Rev. Mol. Diagn. 4, 761–777 (2004). et al.
- Cross-platform comparability of microarray technology: intra-platform consistency and appropriate data analysis procedures are essential. BMC Bioinformatics 6 Suppl 2, S12 (2005). et al.
- The MicroArray Quality Control (MAQC) project shows inter- and intraplatform reproducibility of gene expression measurements. Nat. Biotechnol. 24, 1151–1161 (2006). et al.
- Rat toxicogenomic study reveals analytical consistency across microarray platforms. Nat. Biotechnol. 24, 1162–1169 (2006). et al.
- Evaluation of DNA microarray results with quantitative gene expression platforms. Nat. Biotechnol. 24, 1115–1122 (2006). et al.
- Performance comparison of one-color and two-color platforms within the MicroArray Quality Control (MAQC) project. Nat. Biotechnol. 24, 1140–1150 (2006). et al.
- Using RNA sample titrations to assess microarray platform performance and normalization techniques. Nat. Biotechnol. 24, 1123–1131 (2006). et al.
- Evaluation of external RNA controls for the assessment of microarray performance. Nat. Biotechnol. 24, 1132–1139 (2006). et al.
- Multiple-laboratory comparison of microarray platforms. Nat. Methods 2, 345–350 (2005). et al.
- Arrays of hope. Cell 127, 657–659 (2006).
- Reproducible and reliable microarray results through quality control: good laboratory proficiency and appropriate data analysis practices are essential. Curr. Opin. Biotechnol. 19, 10–18 (2008). , , &
- Comparison of discrimination methods for the classification of tumors using gene expression data. J. Am. Stat. Assoc. 97, 77–87 (2002). , &
- Voluntary exploratory data submissions to the US FDA and the EMA: experience and impact. Nat. Rev. Drug Discov. 9, 435–445 (2010). et al.
- Gene expression profiling predicts clinical outcome of breast cancer. Nature 415, 530–536 (2002). et al.
- Validation and clinical utility of a 70-gene prognostic signature for women with node-negative breast cancer. J. Natl. Cancer Inst. 98, 1183–1192 (2006). et al.
- Interlaboratory performance of a microarray-based gene expression test to determine tissue of origin in poorly differentiated and undifferentiated cancers. J. Mol. Diagn. 10, 67–77 (2008). et al.
- Noninvasive discrimination of rejection in cardiac allograft recipients using gene expression profiling. Am. J. Transplant. 6, 150–160 (2006). et al.
- Microarrays: retracing steps. Nat. Med. 13, 1276–1277, author reply 1277–1278 (2007). , &
- Repeatability of published microarray gene expression analyses. Nat. Genet. 41, 149–155 (2009). et al.
- High-resolution serum proteomic patterns for ovarian cancer detection. Endocr. Relat. Cancer 11, 583–584, author reply 585–587 (2004). , , &
- Selection bias in gene extraction on the basis of microarray gene-expression data. Proc. Natl. Acad. Sci. USA 99, 6562–6566 (2002). &
- Using DNA microarrays for diagnostic and prognostic prediction. Expert Rev. Mol. Diagn. 3, 587–595 (2003).
- Interlaboratory comparability study of cancer gene expression analysis using oligonucleotide microarrays. Clin. Cancer Res. 11, 565–572 (2005). et al.
- Gene expression-based survival prediction in lung adenocarcinoma: a multi-site, blinded validation study. Nat. Med. 14, 822–827 (2008). et al.
- K-nearest neighbors (KNN) models for microarray gene-expression analysis and reliable clinical outcome prediction. Pharmacogenomics J. 10, 292–309 (2010). et al.
- Critical review of published microarray studies for cancer outcome and guidelines on statistical analysis and reporting. J. Natl. Cancer Inst. 99, 147–157 (2007). &
- Prediction of survival in follicular lymphoma based on molecular features of tumor-infiltrating immune cells. N. Engl. J. Med. 351, 2159–2169 (2004). et al.
- Immune signatures in follicular lymphoma. N. Engl. J. Med. 352, 1496–1497, author reply 1496–1497 (2005).
- Functional analysis of multiple genomic signatures demonstrates that classification algorithms choose phenotype-related genes. Pharmacogenomics J. 10, 310–323 (2010). et al.
- That BLUP is a good thing: the estimation of random effects. Stat. Sci. 6, 15–32 (1991).
- Unbiased recursive partitioning: a conditional inference framework. J. Comput. Graph. Statist. 15, 651–674 (2006). , &
- Prognostic gene signatures for non-small-cell lung cancer. Proc. Natl. Acad. Sci. USA 106, 2824–2828 (2009). et al.
- Effect of training sample size and classification difficulty on the accuracy of genomic predictors. Breast Cancer Res. 12, R5 (2010). et al.
- Assessing classifiers from two independent data sets using ROC analysis: a nonparametric approach. IEEE Trans. Pattern Anal. Mach. Intell. 28, 1809–1817 (2006). , &
- On the repeated use of databases for testing incremental improvement of computer-aided detection schemes. Acad. Radiol. 11, 103–105 (2004). , &
- Microarray data analysis: from disarray to consolidation and consensus. Nat. Rev. Genet. 7, 55–65 (2006). , , &
- Classification based upon gene expression data: bias and precision of error rates. Bioinformatics 23, 1363–1370 (2007). , &
- A comparison of batch effect removal methods for enhancement of prediction performance using MAQC-II microarray gene expression data. Pharmacogenomics J. 10, 278–291 (2010). et al.
- Consistency of predictive signature genes and classifiers generated using different microarray platforms. Pharmacogenomics J. 10, 247–257 (2010). et al.
- Genomic indicators in the blood predict drug-induced liver injury. Pharmacogenomics J. 10, 267–277 (2010). et al.
- Comparison of performance of one-color and two-color gene-expression analyses in predicting clinical endpoints of neuroblastoma patients. Pharmacogenomics J. 10, 258–266 (2010). et al.
- Assessing sources of inconsistencies in genotypes and their effects on genome-wide association studies with HapMap samples. Pharmacogenomics J. 10, 364–374 (2010). et al.
- Application of genomic biomarkers to predict increased lung tumor incidence in 2-year rodent cancer bioassays. Toxicol. Sci. 97, 55–64 (2007). , , &
- A gene expression biomarker provides early prediction and mechanistic assessment of hepatic tumor induction by nongenotoxic chemicals. Toxicol. Sci. 99, 90–100 (2007). , &
- Development of a large-scale chemogenomics database to improve drug candidate selection and to understand mechanisms of chemical toxicity and action. J. Biotechnol. 119, 219–244 (2005). et al.
- Gene expression response in target organ and whole blood varies as a function of target organ injury phenotype. Genome Biol. 9, R100 (2008). et al.
- Total RNA yield and microarray gene expression profiles from fine-needle aspiration biopsy and core-needle biopsy samples of breast carcinoma. Cancer 97, 2960–2971 (2003). et al.
- Determination of oestrogen-receptor status and ERBB2 status of breast carcinoma: a gene-expression profiling study. Lancet Oncol. 8, 203–211 (2007). et al.
- Pharmacogenomic predictor of sensitivity to preoperative chemotherapy with paclitaxel and fluorouracil, doxorubicin, and cyclophosphamide in breast cancer. J. Clin. Oncol. 24, 4236–4244 (2006). et al.
- The molecular classification of multiple myeloma. Blood 108, 2020–2028 (2006). et al.
- A validated gene expression model of high-risk multiple myeloma is defined by deregulated expression of genes mapping to chromosome 1. Blood 109, 2276–2284 (2007). et al.
- Thalidomide and hematopoietic-cell transplantation for multiple myeloma. N. Engl. J. Med. 354, 1021–1030 (2006). et al.
- High-risk myeloma: a gene expression based risk-stratification model for newly diagnosed multiple myeloma treated with high-dose therapy is predictive of outcome in relapsed disease treated with single-agent bortezomib or high-dose dexamethasone. Blood 111, 968–969 (2008). , , , &
- Translocation t(4;14) retains prognostic significance even in the setting of high-risk molecular signature. Leukemia 22, 459–461 (2008). , , &
- Prediction of survival in multiple myeloma based on gene expression profiles reveals cell cycle and chromosomal instability signatures in high-risk patients and hyperdiploid signatures in low-risk patients: a study of the Intergroupe Francophone du Myelome. J. Clin. Oncol. 26, 4798–4805 (2008). et al.
- Customized oligonucleotide microarray gene expression-based classification of neuroblastoma patients outperforms current clinical risk stratification. J. Clin. Oncol. 24, 5070–5078 (2006). et al.
The MAQC-II project was funded in part by the FDA's Office of Critical Path Programs (to L.S.). Participants from the National Institutes of Health (NIH) were supported by the Intramural Research Program of NIH, Bethesda, Maryland or the Intramural Research Program of the NIH, National Institute of Environmental Health Sciences (NIEHS), Research Triangle Park, North Carolina. J.F. was supported by the Division of Intramural Research of the NIEHS under contract HHSN273200700046U. Participants from the Johns Hopkins University were supported by grants from the NIH (1R01GM083084-01 and 1R01RR021967-01A2 to R.A.I. and T32GM074906 to M.M.). Participants from the Weill Medical College of Cornell University were partially supported by the Biomedical Informatics Core of the Institutional Clinical and Translational Science Award RFA-RM-07-002. F.C. acknowledges resources from The HRH Prince Alwaleed Bin Talal Bin Abdulaziz Alsaud Institute for Computational Biomedicine and from the David A. Cofrin Center for Biomedical Information at Weill Cornell. The data set from The Hamner Institutes for Health Sciences was supported by a grant from the American Chemistry Council's Long Range Research Initiative. The breast cancer data set was generated with support of grants from NIH (R-01 to L.P.), The Breast Cancer Research Foundation (to L.P. and W.F.S.) and the Faculty Incentive Funds of the University of Texas MD Anderson Cancer Center (to W.F.S.). The data set from the University of Arkansas for Medical Sciences was supported by National Cancer Institute (NCI) PO1 grant CA55819-01A1, NCI R33 Grant CA97513-01, Donna D. and Donald M. Lambert Lebow Fund to Cure Myeloma and Nancy and Steven Grand Foundation. We are grateful to the individuals whose gene expression data were used in this study. All MAQC-II participants freely donated their time and reagents for the completion and analyses of the MAQC-II project. The MAQC-II consortium also thanks R. O'Neill for his encouragement and coordination among FDA Centers on the formation of the RBWG. The MAQC-II consortium gratefully dedicates this work in memory of R.F. Wagner who enthusiastically worked on the MAQC-II project and inspired many of us until he unexpectedly passed away in June 2008.