Dataset Information

Aird: A computation-oriented mass spectrometry data format enables a higher compression ratio and less decoding time

ABSTRACT: We describe "Aird", an opensource and computation-oriented format with controllable precision, flexible indexing strategies, and high compression rate. Aird provides a novel compressor called Zlib-Diff-PforDelta (ZDPD) for m/z data. Compared with Zlib only, m/z data size is about 55% lower in Aird on average. With the high-speed decoding and encoding performance brought by the Single Instruction Multiple Data(SIMD) technology used in the ZDPD, Aird merely takes 33% decoding time compared with Zlib. We used the open dataset HYE, which contains 48 raw files from SCIEX TripleTOF 5600 and TripleTOF6600. The total file size is 206GB as the vendor format. The total size increases to 854GB after converting to mzML with 32-bit encoding precision. While it takes only 189GB when using Aird. Aird uses JavaScript Object Notation (JSON) for metadata storage. Aird-SDK is written in Java and AirdPro is a GUI client for vendor file converting which is written in C#. They are freely available at https://github.com/CSi-Studio/Aird-SDK and https://github.com/CSi-Studio/AirdPro

ORGANISM(S): Homo Sapiens

SUBMITTER: Miaoshan Lu

PROVIDER: PXD025310 | iProX | Mon Apr 12 00:00:00 BST 2021

REPOSITORIES: iProX

ACCESS DATA

Publications

Aird: a computation-oriented mass spectrometry data format enables a higher compression ratio and less decoding time.

Lu Miaoshan M An Shaowei S Wang Ruimin R Wang Jinyin J Yu Changbin C

BMC bioinformatics 20220112 1

<h4>Background</h4>With the precision of the mass spectrometry (MS) going higher, the MS file size increases rapidly. Beyond the widely-used open format mzML, near-lossless or lossless compression algorithms and formats emerged in scenarios with different precision requirements. The data precision is often related to the instrument and subsequent processing algorithms. Unlike storage-oriented formats, which focus more on lossless compression rate, computation-oriented formats concentrate as much ...[more]

PMID: 35021987

Similar Datasets

Project description:Dear Juan Francisco, Dear Pablo we finished analyzing your latest order. In short, the data looks great! Your neg. controls behave as expected and cluster tightly. The probe (JT-195) enriches for a lot of prey proteins. Many are shared with the cysteinome control. Replicates are very similar. Kind regards, Tobias *** Always reply to workflow results by using the order comments function in b-fabric! Please do NOT send emails to proteomics@ or individual people involved in workflow processing (order coach). *** 1 Downloading results packaged in a b-fabric workunits [https://fgcz-bfabric.uzh.ch/wiki/tiki-index.php?page=WorkunitDownload] 2 General data processing logic LC-MS raw data => Dia-NN => proLFQ/SAINTexpress => Interaction proteomics report |-> ScaffoldDIA 3 Workunit guidance 3.1 Dia-NN The main output of Dia-NN is called <WU#>_report.tsv and is a precursor-centric table in long format. It describes chromatographic features mapped to peptides. This file can be loaded in downstream processing tools like DIAgui [https://github.com/mgerault/DIAgui] or amica [https://github.com/tbaccata/amica] or a spread sheet calculator. For direct exploration of Dia-NN output we recommend using the so called matrix files named <WU#>_report.<level>_matrix.tsv These files summarize the main report for pr (precursor), pg (protein group), or gg (gene group) level by the maxLFQ algorithm in matrix-like tables (feature x sample). Be aware that maxLFQ estimates are tailored towards comparisons between experimental conditions (also called horizontal comparisions). They should not be used for rank sorting proteins, as estimator for the abs. amount of proteins. For this special estimators like topN or iBAQ should be used (available in DIAgui) [https://github.com/vdemichev/DiaNN] 3.2 SAINTexpress The main interaction scoring results are summarized in the interaction proteomics report. This report is named SaintExpressReport*_<WU#>.html and can be viewed in any web browser. All numbers shown in the report are extracted from the SAINTexpress output which is available as DIANN_<WU#>_data.xlsx. The list tab summarizes all bait-prey pairs scored by the ML model for the different comparisons. We judge preys by co-filtering in effect size (FoldChange) and significance dimension (BFDR). Suitable cutoffs are very experiment/data dependent, especially in the effect size dimension. We tend to use BFDR <5% and FC >1 as initial thresholds. Refiltering according to known interactions is always a good approach to arrive at meaningful candidate lists. The ORA_Bait_<baitName>_<preyName>.txt files can be used to upload candidate lists to web tools like String-DB [https://string-db.org] for network/enrichment analysis. They are always filtered according to parameters shown in the interaction proteomics report. [https://saint-apms.sourceforge.net/Main.html] 3.3 ScaffoldDIA The main output of protein identification and quantification performed by Dia-NN is also available as ScaffoldDIA formatted file (<WU#>_result.sdia). ScaffoldDIA allows to browse LC-MS datasets using a GUI. The ScaffoldDIA app can be downloaded free of charge from [https://www.proteomesoftware.com/products/scaffold-dia] and runs on all mayor (OS) platforms. Be aware that Dia-NN matrix files might be different to ScaffoldDIA outputs, since ScaffoldDIA takes the Dia-NN main report (which describes chromatographic features mapped to peptides) and recalculates protein inference, reestimate protein quantifies. The output in the app also heavily depends on the set filter options. We do NOT recommend using ScaffoldDIA exports for statistical downstream processing, but only for interactive explorative analysis.

Dataset Information

Aird: A computation-oriented mass spectrometry data format enables a higher compression ratio and less decoding time

Publications

Aird: a computation-oriented mass spectrometry data format enables a higher compression ratio and less decoding time.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets