TECHNICAL RESOURCES · 生信分析

Bioinformatics

Q: Should variant calling use HaplotypeCaller or Mutect2?

GATK HaplotypeCaller is used for germline variants and Mutect2 for somatic variants, followed by GATK VQSR or hard filtering to remove low-quality variants.

Q: STAR or HISAT2 for RNA-seq alignment, and what do I quantify with?

The pipeline recommends STAR, or HISAT2, for alignment to reference genomes; featureCounts or StringTie then produce gene-count matrices.

Q: What happens to my raw MS data before screening for differential proteins?

Thermo RAW / Bruker .d / WIFF files go through a DIA-NN library search or Spectronaut direct-DIA analysis at FDR < 1%, then MaxLFQ or Top3 quantification with log2 transformation; limma or MSstats screens differential proteins.

Q: How are differential metabolites selected?

By t-test plus VIP from PLS-DA, using VIP > 1 and p < 0. 05;

Q: Which files do I receive when the analysis is finished?

Raw files in FASTQ, RAW and WIFF formats with MD5 checksums, a complete PDF analysis report, publication-grade vector graphics (AI/SVG/PDF), editable Excel tables and OmicsHub interactive analysis reports.

Core toolchains across all domains, covering mainstream bioinformatics needs

Methodology note · content is citable as-is

01Toolchain Overview

Core toolchains across all domains, covering mainstream bioinformatics needs

Genomics tools

BWA、GATK、Samtools、ANNOVAR

Transcriptomics tools

STAR、featureCounts、DESeq2、Salmon

Proteomics tools

DIA-NN、Spectronaut、MaxQuant、Philosopher

Metabolomics tools

XCMS、MS-DIAL、MetaboAnalyst、MZmine

02Genomics Analysis Pipeline

Complete pipeline from raw data to variant annotation

Data QC

FastQC + MultiQC summary reports; adapter and low-quality read trimming (Trimmomatic / fastp)

Alignment

BWA-MEM alignment to reference genomes (hg38 / mm10); Samtools sort & index

Variant Calling

GATK HaplotypeCaller (germline) / Mutect2 (somatic)

Variant Filtering

GATK VQSR / hard filtering to remove low-quality variants

Variant Annotation

ANNOVAR / VEP / SnpEff annotation of variant function, frequency and pathogenicity

Summary

VCF files, annotation tables, Circos plots, mutational-signature analysis

03Transcriptomics Analysis Pipeline

From RNA-seq data to differential-expression visualization

Data QC

FastQC + MultiQC; Trimmomatic adapter trimming

Alignment

STAR (recommended) / HISAT2 alignment to reference genomes

Quantification

featureCounts / StringTie gene-count matrices

Differential Expression

DESeq2 / edgeR differential-expression analysis, |log2FC| > 1, FDR < 0.05

Functional Enrichment

clusterProfiler GO/KEGG enrichment; genome-wide GSEA

Visualization

volcano plots, heatmaps, PCA plots, MA plots

04Proteomics Analysis Pipeline

From raw MS data to protein-interaction networks

Raw MS Data

Thermo RAW / Bruker .d / WIFF raw files

Database Search

DIA-NN library search / Spectronaut direct-DIA analysis, FDR < 1%

Protein Quantification

MaxLFQ or Top3 quantification with log2 transformation

Differential-Protein Screening

limma / MSstats，|log2FC| > 0.58, adjusted p < 0.05

Functional Annotation

GO, KEGG, Reactome enrichment; protein-interaction networks (STRING)

Visualization

volcano plots, heatmaps, PCA, PPI network diagrams

05Metabolomics Analysis Pipeline

From peak detection to metabolic-pathway visualization

Peak Detection & Alignment

XCMS / MS-DIAL peak detection and retention-time alignment

Metabolite Identification

MS1 m/z + MS/MS spectral matching (HMDB/METLIN/MassBank)

Preprocessing

missing-value imputation (KNN), PQN normalization, log2 transformation

Differential-Metabolite Screening

t-test + VIP（PLS-DA），VIP > 1, p < 0.05

Pathway Analysis

MetaboAnalyst / KEGG pathway enrichment and topology analysis

Visualization

volcano plots, box plots, heatmaps, pathway bubble charts

06Integration Tools

Tool selection for multi-omics data integration and visualization

Tool / package	Use case	Input data type	Output
mixOmics (R)	Multi-Omics Integration	gene-expression, protein, metabolite matrices	DIABLO models, loading plots
MOFA (R/Python)	unsupervised multi-omics reduction	any multi-omics matrices	factor matrices, variance decomposition
WGCNA (R)	weighted gene co-expression networks	gene-expression matrices	module detection, module–trait association
Cytoscape + stringApp	network visualization & integration	protein/gene lists	interaction networks, enriched modules
iPath / Pathview (R)	pathway visualization	metabolite/gene lists	colored pathway maps

07Analysis-Report Deliverables

Complete data-analysis deliverables meeting publication requirements

Raw Data

raw files in FASTQ, RAW and WIFF formats with MD5 checksums

Analysis Report

complete PDF analysis report with methods and result interpretation

Charts & Tables

publication-grade vector graphics (AI/SVG/PDF) and editable Excel tables

Cloud Platform

OmicsHub interactive analysis reports with online filtering and data download

Q&AFAQ

Should variant calling use HaplotypeCaller or Mutect2?

GATK HaplotypeCaller is used for germline variants and Mutect2 for somatic variants, followed by GATK VQSR or hard filtering to remove low-quality variants.

STAR or HISAT2 for RNA-seq alignment, and what do I quantify with?

The pipeline recommends STAR, or HISAT2, for alignment to reference genomes; featureCounts or StringTie then produce gene-count matrices.

What happens to my raw MS data before screening for differential proteins?

Thermo RAW / Bruker .d / WIFF files go through a DIA-NN library search or Spectronaut direct-DIA analysis at FDR < 1%, then MaxLFQ or Top3 quantification with log2 transformation; limma or MSstats screens differential proteins.

How are differential metabolites selected?

By t-test plus VIP from PLS-DA, using VIP > 1 and p < 0. 05;

Which files do I receive when the analysis is finished?

Raw files in FASTQ, RAW and WIFF formats with MD5 checksums, a complete PDF analysis report, publication-grade vector graphics (AI/SVG/PDF), editable Excel tables and OmicsHub interactive analysis reports.