The ENCODE Imputation Challenge: a critical assessment of methods for cross-cell type imputation of epigenomic profiles

A promising alternative to comprehensively performing genomics experiments is to, instead, perform a subset of experiments and use computational methods to impute the remainder. However, identifying the best imputation methods and what measures meaningfully evaluate performance are open questions. We address these questions by comprehensively analyzing 23 methods from the ENCODE Imputation Challenge. We find that imputation evaluations are challenging and confounded by distributional shifts from differences in data collection and processing over time, the amount of available data, and redundancy among performance measures. Our analyses suggest simple steps for overcoming these issues and promising directions for more robust research.

Automated summary

A promising alternative to conducting comprehensive genomics experiments is to perform a subset of experiments and use computational methods to impute the remaining data. However, determining the best imputation methods and meaningful performance evaluation measures remains an open question. In this study, we address these questions by comprehensively analyzing 23 methods from the ENCODE Imputation Challenge. Our findings indicate that imputation evaluations are challenging due to distributional shifts caused by differences in data collection and processing, the amount of available data, and redundancy among performance measures. Our analyses provide insights into overcoming these challenges and offer promising directions for more robust research.