You shape the dashboard the program runs on

The anti-CD20 program just landed its first round of hits. The program lead's job for the morning is to build the dashboard the team will run their weekly review from: a single page that shows where each candidate stands, how the panel is trending, and what's coming next.

The program lead starts from an empty dashboard and lays in the views the team actually reads off: the program's antibodies as a ranked list, the projects and their phases on a timeline, assay throughput as a trend line, the headline numbers each paired with its change since last week, all sized and placed to match how the team reads the program. That candidate list can simply show the program's antibodies with the columns that matter, or rank them with a scoring profile that re-orders itself as new data lands, so the one view serves equally as a search tool and as a live leaderboard. When the layout is right, the program lead rolls it out org-wide so the whole team works from the same view. 

You compare the round's hits side by side

Round four of the panning campaign produced a couple of dozen hits across multiple plates. The work now is figuring out which ones to take forward, and that means looking at every hit's binding data, sequence, and lineage in one place.

A scientist loads the list of hits side by side in one workspace. One view lays every candidate out as a row and pulls in the data that the decision needs: ELISA reads, BLI kinetics, and framework and CDR annotations, all presented in a single searchable table with built-in plotting capabilities. A companion view aligns the same candidates' heavy or light chains, so the sequence differences sit one toggle from the numbers. The comparison that used to mean three tabs, two exports, and a spreadsheet happens in one place, against live data.

You define how the team scores its leads

The hit comparison narrowed the candidates from a few dozen to a few that look promising. The next move is locking in what the team considers a "good" candidate, weighted by which criteria and what thresholds matter, so the ranking the team works from is the ranking the team agreed to.

A scoring profile is that definition. Its criteria are the things that actually decide the call: binding tight enough, expression high enough, the sequence clean of liabilities. The criteria carry the direction that counts as good, how much each one weighs against the rest, and the goal to measure against. Because a program usually needs coverage as well as potency, the profile can spread the ranking across V-gene families or epitope bins, so the top of the list is a representative panel rather than ten versions of one clone. The profile belongs to the program, scoped to it and to the candidate format it applies to, so from here on the team ranks against one agreed bar instead of a series of spreadsheets in which each scientist weights a little differently.

You watch the ranked list refresh as data lands

The scoring profile is defined. Now the team's working surface for decisions becomes the ranked candidate list, a list they can open every morning to see where the panel stands.

The candidates come back ordered best to worst, each showing its score and how many of the program's goals it clears. A summary across the top says how many clear every goal and how well the panel covers its diversity targets. The list isn't a snapshot: when a fresh ELISA plate or a just-finished Biacore run lands, Affinity re-scores the affected candidates. The "top candidates" are always the current ones, because the list is a live view of the registry through the team's profile.

You build the report for the program review

Friday afternoon. The team's program review is in an hour, and what the director wants to see is the cross-candidate report of every clone the team is tracking, with the binding data, sequences, and potential liabilities on the same row, and a few candidates promoted to the next stage.

The program lead pulls that report in one place, and it spans all the formats that comprise the program: antibodies next to the nanobodies next to all the different bispecific formats, each carrying its own assay columns joined into the same table, the comparison a cross-format program needs and no per-format export can assemble. From the report, the standouts can be added to a named set the team can return to, and the chosen few promoted to the next stage, all without leaving the report.

Why choose Affinity for reporting

• One registry, every analysis surface. Dashboards, the Analyze view, scoring profiles, and report pages all read from the same data the rest of Affinity writes to, so there is no nightly sync, no separate BI stack, and no exports that drift away from the source.

• Analysis you can act on, not just read. Promote the standouts to the next stage and add them to a named set the team can return to. The decision and the record of it happen in one place, without an export-and-re-import round trip.

• Binding data and sequences, one toggle apart. The Analyze view pairs an assay-data report builder with multiple sequence alignment, so the numbers that rank a candidate and the sequence differences that help explain them sit in the same window instead of two tools.
  
  

• Cross-format comparison. Analysis can put antibodies, scFvs, nanobodies, and any other format you design in the same report, useful when a program is evaluating a panel across formats and the comparison columns have to span per-format assay tables.

• Transparent pricing. $175 per user per month, every module included. The Reporting module isn't an add-on tier and the dashboards aren't a separate license.

• Founded in 2011, focused exclusively on biologics R&D — more than a decade of building the analysis layer the scientists who actually use it ask for.