Cross-References: Protein Bioinformatics DNA/RNA Bioinformatics
———————————————————————————————-
DNA Databases BLAST Genbank at NCBI EMBL DNA DDBJ
http://www.ncbi.nlm.nih.gov/LocusLink/” style=”color: rgb(146, 172, 186); -webkit-transition: background-color 0.2s ease-in-out, color 0.2s ease-in-out; transition: background-color 0.2s ease-in-out, color 0.2s ease-in-out;”>LocusLink (NCBI)
Sequence Read Archive Data (SRA): (NIH largest repository of high throughput sequencing data)
NIH (genome)
Annotation on genes: MedMiner
Analysis of complete genomes: PEDANT
Cloning/Enzymes: BRENDA SwissProt Restriction Enzyme Database Cloneit DNA artist download
Cut your DNA Sequence: NEB Cutter Web Cutter
Vector contamination: EMBL EBI NCBI VecScreen
Mammalian Gene Collection: Mammalian Gene Collection (MGI) (cDNA sequences for human, rat & mouse)
Genes involved in Human Disease: GeneCards OMIN
Open Reading Frames (OPFs): ORF Finder ORF Finder (NCBI) Expasy-translate Sequence Manip Site
Comparison of genomes: Institute for Genomic Research TIGR Vista PipMaker
Sequence Submission: GenBank Submissions (BankIt) NCBI GenBank Examples NCBI Sequin (This is software developed by NCBI which must be downloaded. Use this for sequences which are too short to submit online with BankIt or where you want more control over your submissions) Sequin Help Sequin Factshttp://www.ebi.ac.uk/embl/Submission/webin.html” style=”color: rgb(146, 172, 186); -webkit-transition: background-color 0.2s ease-in-out, color 0.2s ease-in-out; transition: background-color 0.2s ease-in-out, color 0.2s ease-in-out;”>EMBL (WebIn)
Sequence Analysis: Metagene
Sequence Conversion: (Convert sequence files from one particular format to another format)
ReadSeq Readseqimple ReadseqBaylor Readseqfinice Readseqbioportal readseqebi Redseqbimas
Specific genomes: FlyBase newsnetwork
Wiring Diagrams of life: KEGG
Deep learning models: AlphaGenome (Google Deep Mind that can take as input 1 Mb of DNA and predict thousands of functional genomic tracks up to single-base pair resolution across diverse modalities. DeepMind has made AlphaGenome available for non-commercial research. )
Introduction:
Information used to predict genes includes signals in the sequence, content statistics and similarity to known genes. As noted by Attwood (Science, 290, 2000), there are many obstacles to accurate gene counting. First there is the problem of what exactly constitutes a gene. Is it a heritable unit corresponding to an observable phenotype? Is it a packet of genetic information that encodes an RNA or a protein? Must it be translated? Answers to these question affect estimates of the total number of genes in sequenced genomes.
A long-standing goal in genetics is to accurately predict the effect of modifying each of the three billion nucleotides in the human genome with respect to gene-regulatory activity, ranging from chromatin accessbility and transcriptional activaiton to splicing and polyadenylation. Machin-learning models trained to predict funciton from DNA sequences have been successful at charectarizing regulatory syntax and interpreting genetic variant effects. Borzoi learns to predict sequencing coverage from a vast set of RNA-seq experiments. It enables variant scoring and itnerpretation through multiple layers of regulation, including transcirption, splicing and polyadenylation. (Linker, “Predicting RNA-seq coverage from DNA sequence as a unifying model of gene regulation, 2023, Nature genetics),
AlphaGenome (see above) is a deep learning model designed to learn the sequence basis of diverse molecular phenotypes form human and mouse DNA. It simultaneously predicts modalities cvoering gene expression (RNA-seq. CAGE and PRO-cap), detailed splicing patterns (splice sites, splice site usage and splice junctions), chromatic state (DNase, ATAC-seq, histone modifications and transcription factor binding) and chromatin contact maps. These span a vareity of biologcial contexts, such as different tissue types, cell types and cell lines).