LambdaRank has proven effective for optimizing information retrieval metrics such as Normalized Discounted Cumulative Gain (NDCG). However, its application to Precision at document k (P@k) poses significant challenges because of the metric’s unique definition, which heavily restricts the number of effective training document pairs. This limitation diminishes the learning signal for relevant documents beyond the top k, potentially resulting in suboptimal performance. To overcome this, we propose LambdaGap, a ranking algorithm inspired by LambdaRank specifically tailored for optimizing P@k. LambdaGap replaces the pairwise weighting scheme in LambdaRank by one where pairs of documents within k positions in the ranking are masked out. We establish a theoretical link between LambdaGap and P@k by identifying the implicit metric optimized by the model. Furthermore, we introduce a new metric, Average Relevance Position beyond document k, which can be used in conjunction with LambdaRank to indirectly optimize for P@k. Our extensive experiments on publicly available datasets demonstrate the effectiveness of the proposed methods, yielding statistically significant improvements in P@k performance and highlighting their potential for more efficient training.

SUMMARY: More than 150 peptide therapeutics are globally in clinical development. Many enzymatic barriers should be crossed by a successful drug to be prosperous in such a process. Therefore, the new peptide drugs must be designed preventing the potential protease cleavage to make the compound less susceptible to protease reaction. We present a new data analysis tool developed in WebMetabase, an approach that stores the information from liquid chromatography mass spectrometry-based experimental data or from external sources such as the MEROPS database. The tool is a chemically aware system where each peptide substrate is presented as a sequence of structural blocks (SBs) connected by amide bonds and not being limited to the natural amino acids. Each SB is characterized by its pharmacophoric and physicochemical properties including a similarity score that describes likelihood between a SB and each one of the other SBs in the database. This methodology can be used to perform a frequency analysis to discover the most frequent cleavage sites for similar amide bonds, defined based on the similarity of the SB that participate in such a bond within the experimentally derived and/or public database. AVAILABILITY AND IMPLEMENTATION: http://webmetabase.com:8182/WebMetabaseBioinformatics/. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.