Learning center

Protein composition: exact counts, rounded percentages and unavailable predictions

Explore an explicit protein sequence, inspect residue counts and interpret composition, entropy and mass summaries. A missing solubility result is not an estimated value.

Explore an explicit protein sequence, inspect residue counts and interpret composition, entropy and mass summaries. A missing solubility result is not an estimated value.

1. Supply one explicit protein sequence

The saved input is ACTB_protein.fasta, the deposited NP_001092.1 protein associated with ACTB transcript NM_001101.5. It contains 375 amino-acid residues. The guide’s evidence inventory pins the FASTA and the saved deposited-reference fixture. This is the protein sequence, not the 1,812-nucleotide transcript and not a 50-residue display preview.

Load the protein FASTA and open Protein Properties in Bioinformatics. The current contract requires one nonempty sequence using the standard 20 amino-acid symbols. It normalizes letter case and whitespace. One terminal stop symbol may be removed with a warning; internal stops and unsupported residue symbols are refused. Input identified as nucleotide-only DNA/RNA is not a protein input. Preserve the source sequence and any normalization notice.

The current input contract uses the first selected sequence record or the explicit sequence field. Do not assume a multi-record FASTA produces a pooled composition or a separate report for every record. Confirm the selected protein and the QC sequence length before running. The saved example has QC pass, length 375 and no warnings.

2. Select the supported question and retain unavailable outcomes

The interface has twenty initially selected analysis tracks. For the recorded scope, clear the other selections and retain AA Composition and Solubility Prediction (unavailable). These correspond to amino_acid_composition and protein_solubility. The second selection demonstrates a disclosed limitation; it does not request an implemented prediction.

The earlier native run completed with one unavailable method. It produced a composition figure/table, a solubility availability figure/table, an input-QC table and a summary. An unavailable panel appearing among exported results is not a successful solubility estimate. The current source still states that no validated estimator with defined host, expression conditions and model provenance is implemented.

The saved solubility CSV has the generic status token not_uploaded. Read its Reason: the estimator is not implemented. The protein itself was loaded, passed QC and supplied the composition result. Do not interpret this token as a missing protein upload, a solubility score of zero or evidence that ACTB is insoluble.

3. Keep exact counts separate from printed percentages

The saved composition table retains all twenty residue counts. They sum to 375. Each residue percentage is displayed with two decimal places: alanine A is 29 (7.73%), cysteine C is 6 (1.60%), and tryptophan W is 4 (1.07%). Use the integer count and declared denominator when exact proportions matter. Rounded percentage strings need not sum to exactly 100%.

The composition plot’s uniform reference is 5% per residue, corresponding to twenty equally represented symbols. It is a comparison line, not an expected ACTB composition, a confidence limit or a biological target. The earlier independent receipt checked all twenty counts against the deposited protein. This guide reopens the saved fields, checks their totals and printed percentage formatting, and preserves that receipt; it does not run a new protein analysis.

4. Read each residue group’s definition

Saved groupMembers and result
NonpolarGAVLIPFMW: 187 residues (49.9%)
Polar unchargedSTNQCY: 93 residues (24.8%)
PositiveKRH: 46 residues (12.3%)
NegativeDE: 49 residues (13.1%)
AromaticFWY: 32 residues (8.5%)

The first four groups partition the twenty-symbol alphabet under this method. Aromatic, tiny and small groups overlap those categories. Adding every group count together would count some residues more than once. The labels “positive” and “negative” describe residue categories here, not a measured net charge at a specified pH.

The separate QC field Hydrophobic fraction uses AILMFWV, a different set from the composition table’s nonpolar group. It counts 140 of 375 residues: the QC CSV shows 0.3733, while its JSON retains 0.37333333333333335. The nonpolar total additionally includes glycine and proline. A difference between 37.33% and 49.9% therefore has a documented definition; do not silently rename one group to match the other.

5. Limit entropy and mass interpretations

The table reports composition Shannon entropy 2.9004, maximum ln(20) = 2.9957 and normalized complexity 0.9682. The source uses natural logarithms, so the unnormalized entropy is in nats and the ratio is dimensionless. These saved strings have four decimal places. Composition entropy depends on residue frequencies; it does not retain residue order, identify a fold or validate protein function.

The QC table also carries an estimated molecular weight of 41,735.31 Da. Current QC source sums its amino-acid mass table and subtracts water for peptide-bond formation. This is a sequence-based estimate, not an experimental mass measurement, and the composition receipt does not independently validate that mass. It also differs in definition from the rough 110 Da-per-residue estimate described in the ACTB ORF guide.

Preserve the analyzed sequence, termini and declared assumptions when interpreting any property. The primary ProtParam documentation likewise describes sequence-derived parameters and explicitly limits what can be specified about modifications and molecular assemblies. Reading that reference is not a new comparison against ProtParam. This two-route example does not establish molecular-weight, pI, disorder, secondary-structure, host-expression or other prediction accuracy.

6. Verify complete saved fields and their precision

  1. Open the result viewer and use Export All for the selected results. Keep the input-QC output and the unavailable-model explanation along with the successful composition output.
  2. The reviewed export set contains seven artifacts: two PNGs, three CSVs, summary text and an analysis-report JSON. Its separate artifact manifest records hashes and byte counts for those seven files. The receipt identifies the files ending in _2; earlier exports remain preserved. Identify the set by its manifest rather than assuming a suffix means a different scientific method.
  3. The composition CSV has 33 parameter rows, including all twenty residue counts. The QC CSV has ten rows and the solubility CSV has four. This guide compares all 47 fields with the saved native tables and verifies all seven artifact identities.
  4. Complete field export preserves the values prepared by the app, including formatted strings. Residue percentages have two decimals; group percentages have one; entropy and normalized complexity have four. Export All does not restore hidden precision to those strings. The separate QC JSON contains numeric fractions, but it is not one of the seven listed viewer-export artifacts.
  5. The viewer’s analysis report lists six displayed results: two plots, three tables and one text result. It exports no sequence object in this example. Keep the original FASTA separately; the report and PNGs do not constitute a reloadable scientific session.

The historical QC CSV includes a local path to its separately saved QC JSON. A copied CSV does not make that original machine path portable. Retain the QC JSON as a separately identified companion when needed and inspect paths or other identifying content before sharing an export.

Worked examples describe particular inputs, methods and recorded results. Check current availability and input requirements before planning an evaluation.

← Back to the learning center