Dealing with missing values in proteomics data

Kong, Weijia; Hui, Harvard Wai Hann; Peng, Hui; Goh, Wilson Wen Bin

doi:10.1002/pmic.202200092

Cited by 38 publications

(42 citation statements)

References 132 publications

(226 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Protein quantification is then achieved by summarizing intensities of constituent peptides. Despite advances in many aspects of MS-based proteomics, both instrumental and methodological, missing values in the peptide intensity quantifications remain a common feature of such datasets and present one of most challenging problems for downstream analysis ( Webb-Robertson et al 2015 ; Kong et al 2022 ).…”

Section: Introductionmentioning

confidence: 99%

Neither random nor censored: estimating intensity-dependent probabilities for missing values in label-free proteomics

Smyth

2023

Bioinformatics

View full text Add to dashboard Cite

Motivation Mass spectrometry proteomics is a powerful tool in biomedical research but its usefulness is limited by the frequent occurrence of missing values in peptides that cannot be reliably quantified (detected) for particular samples. Many analysis strategies have been proposed for missing values where the discussion often focuses on distinguishing whether values are missing completely at random (MCAR), missing at random (MAR) or missing not at random (MNAR). Results Statistical models and algorithms are proposed for estimating the detection probabilities and for evaluating how much statistical information can or cannot be recovered from the missing value pattern. The probability that an intensity is detected is shown to be accurately modeled as a logit-linear function of the underlying intensity, showing that missing value process is intermediate between MAR and censoring. The detection probability asymptotes to 100% for high intensities, showing that missing values unrelated to intensity are rare. The rule applies globally to each dataset and is appropriate for both high and lowly expressed peptides. A probability model is developed that allows the distribution of unobserved intensities to be inferred from the observed values. The detection probability model is incorporated into a likelihood-based approach for assessing differential expression and successfully recovers statistical power compared to omitting the missing values from the analysis. By contrast, imputation methods are shown to perform poorly, either reducing statistical power or increasing the false discovery rate to unacceptable levels. Availability Data and code to reproduce the results shown in this article are available from https://mengbo-li.github.io/protDP/. Supplementary information Supplementary data are available at Bioinformatics online.

show abstract

Section: Introductionmentioning

confidence: 99%

Neither random nor censored: estimating intensity-dependent probabilities for missing values in label-free proteomics

Smyth

2023

Bioinformatics

View full text Add to dashboard Cite

show abstract

“…The hurdles associated with missing values are a recurring issue in data analysis and concern a wide range of fields and applications . The handling of missing values in mass spectrometry (MS)-based proteomics is still actively debated. − While elucidating the best computational approaches to manage missing values is ongoing, the field continues pushing the boundaries of low input acquisitions. Recent technical advances in MS have paved the way for MS-based single-cell proteomics (SCP), − but handling missing values in SCP data is still a clear challenge to principled data analysis .…”

Section: Introductionmentioning

confidence: 99%

Revisiting the Thorny Issue of Missing Values in Single-Cell Proteomics

Vanderaa

Gatto

2023

J. Proteome Res.

View full text Add to dashboard Cite

Missing values are a notable challenge when analyzing mass spectrometry-based proteomics data. While the field is still actively debating the best practices, the challenge increased with the emergence of mass spectrometrybased single-cell proteomics and the dramatic increase in missing values. A popular approach to deal with missing values is to perform imputation. Imputation has several drawbacks for which alternatives exist, but currently, imputation is still a practical solution widely adopted in single-cell proteomics data analysis. This perspective discusses the advantages and drawbacks of imputation. We also highlight 5 main challenges linked to missing value management in single-cell proteomics. Future developments should aim to solve these challenges, whether it is through imputation or data modeling. The perspective concludes with recommendations for reporting missing values, for reporting methods that deal with missing values, and for proper encoding of missing values.

show abstract

“…In some cases, missingness is meaningful and should not be imputed, while in other cases it may be due to limited instrument sensitivity or algorithmic/statistical issues. The mechanism behind MVs is important in determining the appropriate handling procedure 6 . MVs can be broadly categorized into three types: Missing not at random (MNAR), missing at random (MAR), and missing completely at random (MCAR) 7 .…”

Section: Introductionmentioning

confidence: 99%

“…These methods can range from simple mean or median imputation to more sophisticated techniques such as multiple imputation, hot deck imputation, expectation-maximization algorithm, and various machine learning models such as k-nearest neighbors, Random Forest, and neural networks. The choice of the MVI technique will depend on the characteristics of the data, the proportion of missing values, the type of missingness and the downstream analysis(6).…”

Section: Introductionmentioning

confidence: 99%

“…Some researchers may choose to avoid MVI as much as possible and instead opt for methods that do not require imputation (13). Ultimately, the choice of MVI technique will depend on the characteristics of the data, the proportion of missing values, the type of missingness, and the downstream analysis (6). BEs are another common yet complex problem in high-throughput omics data, similar to MVs (1,14).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Uncovering the consequences of batch effect associated missing values in omics data analysis

Hui

Goh

2023

Preprint

Self Cite

View full text Add to dashboard Cite

Statistical analyses in high-dimensional omics data are often hampered by the presence of batch effects (BEs) and missing values (MVs), but the interaction between these two issues is not well-studied nor understood. MVs may manifest as a BE when their proportions differ across batches. These are termed as Batch-Effect Associated Missing values (BEAMs). We hypothesized that BEAMs in data may introduce bias which can impede the performance of missing value imputation (MVI). To test this, we simulated data with two batches, then introduced over 100 iterations, either 20% and 40% MVs in each batch (BEAMs) or 30% in both (control). K-nearest neighbours (KNN) was then used to perform MVI, in a typical global approach (M1) and a supposed superior batch-sensitized approach (M2). BEs were then corrected using ComBat. The effectiveness of the MVI was evaluated by its imputation accuracy and true and false positive rates. Notably, when BEAMs existed, M2 was generally undesirable as the differing application of MV filtering in M1 and M2 strategies resulted in an overall coverage deficiency. Additionally, both M1 and M2 strategies suffered in the presence of BEAMs, highlighting the need for a novel approach to handle MVI in data with BEAMs.

show abstract

Dealing with missing values in proteomics data

Cited by 38 publications

References 132 publications

Neither random nor censored: estimating intensity-dependent probabilities for missing values in label-free proteomics

Neither random nor censored: estimating intensity-dependent probabilities for missing values in label-free proteomics

Revisiting the Thorny Issue of Missing Values in Single-Cell Proteomics

Uncovering the consequences of batch effect associated missing values in omics data analysis

Contact Info

Product

Resources

About